MacCara 2001

Methods	ITS with monthly data points
Participants	Nova Scotia Seniors' Pharmacare Program: 1995/96: n=107,827, 1996/97: n=10,707, 1997/98: n=107,668
Interventions	Indications had to meet specific criteria for prescription to be filled for reimbursement for fluroquinolones
Outcomes	Drug use, drug costs
Notes	Included 1‐12 month periods prior to policy (Dec 1994‐Nov 1996), Policy implementation Dec 1‐31 1996. On march 1 1998 policy changed and physicians could write specific indication and criteria on prescription slip, Transition period Dec 1, 96 to Feb 28 '97, Post policy analysis to Dec 31 1998. Analysis is confounded by simultaneous publicity campaign on antibiotic resistance.
Risk of bias
Bias	Authors' judgement	Support for judgement
Incomplete outcome data (attrition bias) All outcomes	Low risk	Any undocumented difference in the proportion of missing data in the administrative datasets pre‐ and post‐intervention is unlikely to overturn study results
Selective reporting (reporting bias)	High risk
Other bias	High risk	Seniors who opted to pay out‐of‐pocket for a fluoroquinolone were not included in the study
Knowledge of allocated interventions adequately prevented during the study	Low risk	Data objective: the administrative drug claims database of the Nova Scotia Senior's Pharmacare Program (NSSPP)
Intervention independent of other changes	High risk	A coalition of medical organizations started a national campaign to combat the misuse of antibiotics and heighten the awareness of the Canadian public to the dangers of antibiotic resistance within weeks of the policy introduction. Seasonal bias may also have occurred pre and post as the policy implementation date was December 1st when infectious diseases are on the rise going into winter.
The shape of intervention pre‐specified?	Low risk	The point of analysis is the point of intervention; ie, the date the policy intervention was implemented was used to delineate pre and post policy time periods with adequate data points to capture the shape of the pattern of intervention effect over time
Intervention unlikely to affect data collection?	Low risk	Sources and methods of data collection were the same before and after the intervention