Mason 1998/99.

Methods	Study design: ITS
Participants	Physicians Clinical speciality: general practice/family medicine Level of training: fully trained Setting/country: general practice/UK
Interventions	The PEM studied in this report was an "Effective Health Care" bulletin questioning the cost effectiveness of prescribing SSRIs was distributed to all GPs by the chief medical officer. Original distribution of the bulletin to all GPs occurred in March 1993. We examined the effect of this intervention on prescribing in English primary care using time‐series analysis
Outcomes	2 process outcomes: prescription of SSRIs prescription of tricyclic antidepressants
Notes	‐
Risk of bias
Bias	Authors' judgement	Support for judgement
Intervention independent of other changes ‐ ITS	Unclear risk	Quote, pg. 122: "the Effective Health Care Bulletin, and related article in the BMJ published at the same time, were the first scientific reports to question the widespread switch to SSRIs.  These sparked considerable interest in the media, and also considerable activity from medical and pharmaceutical advisors in the NHS"
Shape of Intervention effect pre‐specified ‐ ITS	Unclear risk	A specific null hypothesis is not provided. Quote pg. 120: "we examined the effect of this intervention on prescribing in English primary care using time series analysis"
Intervention unlikely to affect data collection ‐ ITS	Low risk	The Effective Health Care Bulletin (the intervention) did not affect the data source (Prescriptions Pricing Authority) 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. 120: "these data reflect the total number of prescriptions reimbursed for antidepressants on a quarterly basis"  COMMENT: if a patient does not seek or receive reimbursement, this data could be missed, but this is unlikely to be affected by the publication of the PEMs
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