Summary of findings 3. Educational policy: mailed educational materials on prescribing for physicians and HMO members compared to no educational intervention.

Educational policy: educational materials on prescribing for physicians and HMO members compared to no educational intervention
Patient or population: prescribers or patients that are subjected to regulatory and educational interventions in prescribing due to government, non‐government or third party payer policies Setting: USA Intervention: mailed educational materials on prescribing for physicians and HMO members (HMO B) Comparison: no educational policy intervention (HMO D)
Outcomes	Impact	No. of participants (studies)	Certainty of the evidence (GRADE)
Medicine use	We are uncertain about the effects on medicine use of an educational policy involving educational materials on prescribing mailed to physicians and HMO members, compared to no intervention The share of prescribing for the preferred drug (fexofenadine) increased by 9.5% (95% CI 7.9% to 11.1%) in the HMO that implemented educational materials for physicians and HMO members compared to an increase of 1.6 % (95% CI 0.4% to 2.8%) in the control HMO The share of prescribing for the less preferred drug (loratadine) decreased by −12.5% (95% CI −14.6% to −10.4%) in the HMO that implemented educational materials for physicians and HMO members compared to a decrease of −3.6% (95% CI −5.2% to −2.0%) in the control HMO	1 CBAA	⊕⊖⊖⊖ Very lowB
Health care utilization	Not reported	‐	‐
Health outcomes	Not reported	‐	‐
Costs	The average cost per prescription for the antihistamines increased from USD 38.79 to USD 41.98 or 2% in the educational policy intervention group. The average cost per prescription increased 4.2% in the control group.	1 CBAA	Not assessed
GRADE Working Group grades of evidence High quality: further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: we are very uncertain about the estimate.

^A (Benedetto 2000)

^B Downgraded due to high risk of bias and indirectness (the inclusion of a single study from a high income country)