Skip to main content


Effects of decision aids on the accuracy of patient knowledge of outcome probabilities: an investigation of sources of heterogeneity

Date and Location




Saturday 21 September 2013 - 10:30 - 12:00


Presenting author and contact person

Presenting author

Stephen Gentles

Contact person

Stephen Gentles
Abstract text
Background: Interpretation of prior meta-analysis of randomized controlled trials (RCTs) evaluating patient decision aid effects on the accuracy of knowledge of outcome probabilities is complicated by high, unexplained heterogeneity. Objectives: To explore effect modification from three possible sources of heterogeneity: the type of control intervention, decision aid quality, and patients’ baseline knowledge of probabilities. Methods: A sub-analysis of studies identified in the 2011 Cochrane review on decision aids for people facing treatment and screening decisions was conducted. Additional unpublished data were requested from relevant study authors to maximize the number of eligible studies. RCTs (to 2009) comparing decision aids with standardized probability information to control interventions (lacking such information), and assessing accuracy of patient knowledge of outcome probabilities were included. Proportions of patients in each group with accurate knowledge of outcome probabilities were converted to relative effect measures. Instrument quality was assessed using the IPDAS instrument. Results: Main effects analysis of 17 eligible studies confirmed decision aids significantly improve accuracy of patient knowledge of outcome probabilities (RR=1.80 [1.51, 2.16]), with substantial heterogeneity (87%). Meta-regression indicated the control event rate (CER, reflecting baseline knowledge) is a significant effect modifier (p = 0.001), with over half the variability in ln(OR) explained by the linear relationship with logit Control (R2 = 0.52); this relationship was slightly strengthened after correcting for dependence of the effect measure on CER. The decision aid quality measure suggests potential effect modification (p = 0.037) accounting for some variability (R2 = 0.28). Conclusions: Patients’ baseline rate of knowledge of outcome probabilities is an important variable explaining heterogeneity of decision aid effects on improving accuracy of this knowledge, with greater relative effects observed when patients’ baseline rate of knowledge is low. This may indicate that decision aids are most effective in populations with low awareness.