Background: The impact of providing information about medicines to patients on decisions about starting or continuing them is unknown.
Purpose: To estimate the impact on decisions to start or continue medicines, of providing information to patients about possible benefits and/or harms.
Data sources: Electronic searches from 1980 to October 2010; reference and citation searches of included studies.
Study selection: Two investigators assessed studies' eligibility against inclusion criteria: randomized or pseudorandomized trials; participants older than 16 years and deciding for themselves; one group received information about possible benefits and/or harms of a potentially beneficial medicine, compared with another who did not; d) a measure of decision about starting or continuing a medicine.
Data extraction: One investigator extracted all data, checked by a second.
Data synthesis: Eight studies were included, covering a range of medicines. There was no consistent impact of interventions on decisions about whether to start or continue medicines (pooled odds ratio 1.08; 95% confidence interval [CI], 0.69-1.70; P = 0.74). Among those who received more information, knowledge was increased (pooled mean difference 8.47; 95% CI 2.17-14.77; P = 0.008), and decisional conflict was reduced (pooled mean difference -0.15; 95% CI -0.24 to -0.06; P = .001).
Limitations: A small number of studies across different clinical contexts, of uncertain heterogeneity, were included.
Conclusions: Providing information to patients about possible benefits and/or harms has no consistent effect on the number who decide to start or continue medicines, although it increases patients' knowledge and reduces their decisional conflict.