Adherence intervention research: what have we learned and what do we do next?

Although there is general agreement from studies demonstrating that adherence to inhaled corticosteroid therapy is often inadequate to establish consistent control, relatively little concurrence exists in reports of interventions to correct the problem. Half of the studies reviewed found that the experimental intervention did not change adherence, and behavior change reported by patients was often not accompanied by changes in treatment success. Studies used a variety of methods that differed in quality with findings that were often contradictory. Key limitations in many studies included reliance on inadequate adherence measures, inclusion of convenience samples of well-motivated patients, and assessments of intervention outcomes artificially boosted by attrition of least adherent participants. Research is encouraged into innovative interventions that are brief, easily implemented, and can be tailored to individual patients and diverse clinical settings. Of particular importance is inclusion of hard-to-reach patients, including urban and rural poor and the use of valid measures of adherence at intervals sufficient to establish enduring benefit.