Objective: Attempts to synthesize the evidence on the effects of decision aids have been hampered by the lack of consensus regarding how such effectiveness should be measured. This paper seeks to describe and critically assess the range of measures of effectiveness used in randomized controlled trials of decision aids.
Search strategy: The published systematic reviews of the field were used to identify primary studies evaluating the effects of decision aids.
Inclusion criteria: Non-randomized trials were excluded from this review. As were abstracts and theses of subsequently published studies, methodological papers and reports of subgroups of a study's main publication.
Main results: A wide range of measures were used to evaluate the effectiveness of decision aids. The most commonly used measures sought to assess treatment decisions, patient's knowledge and the decision-making process. This pattern was repeated when primary measures of effectiveness were examined. No study attempted to measure the extent to which decisions made were consistent with patient's values.
Conclusions: Within the current literature there is little consensus on what the aims of decision aids should be. If we can agree that the aim of a decision aid is to help patients make specific personal treatment choices, then evaluations of decision aids should measure the primary effectiveness of their interventions in terms of the extent to which they enable patient's to undergo treatments that agree with their values.