Austin 2005.

Methods	Study design: ITS
Participants	Physicians Clinical speciality: not clear Level of training: fully trained Setting/country: not clear/Canada
Interventions	2 PEMs are studied in this report: the REVERSAL trial, published on 3 March 2004, which demonstrated that for patients with CHD, intensive lipid‐lowering therapy reduced progression of coronary atherosclerosis compared with moderate therapy. 1 month later, the PROVE IT–TIMI22 trial (published on 8 April 2004) demonstrated that among patients who have recently had an ACS, an intensive lipid‐lowering statin regimen provided greater protection against death or major cardiovascular events than did a standard regimen. In both trials, standard therapy consisted of 40 mg/day of pravastatin, whereas intensive therapy consisted of 80 mg/day of atorvastatin. We compared the data before the 2 publications to the data after the 2 publications
Outcomes	2 process outcomes (prescribing): total number of prescriptions of atorvastatin 80 mg/day for residents age 65 years and older in Ontario, Canada total number of prescriptions of pravastatin 40 mg/day for residents age 65 years and older in Ontario, Canada
Notes	We looked at the combined effect of the 2 PEMs because of a lack of data to look at them separately. In this case, the 2 PEMs studied had similar characteristics, and we considered them as a whole (i.e. 1 PEM)
Risk of bias
Bias	Authors' judgement	Support for judgement
Intervention independent of other changes ‐ ITS	Unclear risk	Quote, pg. 1300: "we were unable to account for temporal influences beyond the publication of the results of the trials. In particular, we were unable to account for changes in drug company promotion patterns." Quote, pg. 1300: "because of  the study design and the relatively low monthly number of incident statin users, we were unable to definitively determine whether the trends that we observed were a result of an increase in the number of incident statin users who were being placed on high‐dose atorvastatin or whether they were because of prevalent statin users’ changing therapy"
Shape of Intervention effect pre‐specified ‐ ITS	Unclear risk	Quote, pg. 1297: "the objective of the present study was to examine the impact of the publication of these 2 trials on trends in intensive versus moderate statin therapy in the province of Ontario" COMMENT: the authors do not specify what the expected impact of the intervention is
Intervention unlikely to affect data collection ‐ ITS	Low risk	Quote pg. 1297: "we studied claims for statins to Ontario's universal Drug Benefit program for seniors (ODB) between June 1, 1997 (the month atorvastatin was added to the ODB formulary), and September 30, 2004. The ODB tracks medication use by all 1.4 million residents of Ontario older than 65 years"  COMMENT: data source and method of collection unchanged throughout study
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	The authors used the complete database of all prescription on Ontario, so there is no missing data
Selective reporting (reporting bias) ‐ ITS	Low risk	All relevant outcomes in the methods section were reported in the results section
Other bias ‐ ITS	Low risk	There was no evidence of other risks of bias