Majumdar 2003.

Methods	Study design: ITS
Participants	Physicians Clinical speciality: not clear Level of training: fully trained Setting/country: not clear/US and Canada
Interventions	2 PEMs were studied in this report. The HOPE study demonstrated a 22% reduction in cardiovascular morbidity and mortality, and provided a new indication for ramipril. RALES compared spironolactone with placebo in patients with heart failure and demonstrated a 30% reduction in mortality
Outcomes	4 process outcomes: prescribing patterns of ramipril (in Canada) before and after publication of HOPE prescribing patterns of ramipril (in US) before and after publication of HOPE prescribing patterns of spironolactone (in Canada) before and after publication of RALES prescribing patterns of spironolactone (in US) before and after publication of RALES
Notes	‐
Risk of bias
Bias	Authors' judgement	Support for judgement
Intervention independent of other changes ‐ ITS	Low risk	Quote, pg. 468: "To adjust for potential differences between Canadian and United States physicians in the adoption of published evidence, we examined the effect of the Randomized Aldactone Evaluation Study (RALES) on prescribing trends for spironolactone. This study compared spironolactone with placebo in patients with heart failure and demonstrated a 30% reduction in mortality. RALES was prereleased and published in the same year and the same journal as the HOPE study. Because spironolactone was not promoted by the pharmaceutical industry in either country, any observed differences in prescribing trends should be attributable mostly to a publication effect"
Shape of Intervention effect pre‐specified ‐ ITS	Low risk	Quote, pg. 468: "Therefore, we compared the prescribing trends for ramipril in Canada and the United States to test the hypotheses that publication of the HOPE study would increase the use of ramipril in both countries (publication effect), and that this increase would be greater in Canada (promotion effect)"
Intervention unlikely to affect data collection ‐ ITS	Low risk	The interventions studied (HOPE; RALES) did not affect either the source or the method of data collection
Blinding of outcome assessors (detection bias) ‐ ITS All outcomes	Low risk	The outcome was objective
Incomplete outcome data (attrition bias) ‐ ITS All outcomes	Low risk	Quote, pg. 468: "We used nationally representative drug dispensing information collected by IMS Health (IMS Health‐Canada and IMS Health‐America), which conducts research on prescribing patterns. Methods for data collection are identical in Canada and the United States. The IMS "CompuScript" database collects monthly dispensing records from a representative sample of retail pharmacies. The sample is drawn from 4800 pharmacies in Canada and 51,355 pharmacies in the United States, about two thirds of retail pharmacies"  COMMENT: missing data, if any, were likely to be similar pre‐ and post‐intervention
Selective reporting (reporting bias) ‐ ITS	Low risk	All relevant outcomes in the methods section were reported in results section
Other bias ‐ ITS	Low risk	There was no evidence of other risks of bias