Table A6: Risk of Bias Among Observational Trials for the Impact of eTools.

Author, Year	Appropriate Eligibility Criteria	Appropriate Measurement of Exposure	Appropriate Measurement of Outcome	Adequate Control for Confounding	Complete Follow-Up
Branger et al, 1999 (32)	Serious limitationsa	No limitations	No limitations	No limitations	No limitations
Cebul et al, 2011 (38)	Serious limitationsb	No limitations	No limitations	No limitationsc	No limitations
Crosson et al, 2012 (39)	Serious limitationsb	No limitations	No limitations	No limitationsc	Serious limitationsd
Henderson et al, 2010 (36)	Serious limitationsb	No limitations	No limitations	No limitationsc	No limitationse
Herrin et al, 2012 (40)	Serious limitationsf	No limitations	No limitations	No limitationsc	No limitationsg
Walsh et al, 2012 (41)	Serious limitationsb	No limitations	No limitations	No limitationsc	No limitations
Wells et al, 1996 (42)	Serious limitationsb	No limitations	No limitations	No limitations	Serious limitationsd

Abbreviation: EHR, electronic health record; EMR, electronic medical record; eTools, electronic tools.

^aPhysicians with greatest number of referrals were provided with the electronic intervention, while the others were considered the control group.

^bPhysicians self-selected to use EMRs (or other electronic intervention) and therefore may inherently be different from those who did not.

^cStatistical modelling was applied to adjust for known or otherwise potential confounding factors.

^dIntervention was implemented at the level of physician practice, and this resulted in some flux of individual patients within both study groups.

^eAssessment was conducted at the level of patient encounter; individual patients were not accounted for.

^fUnknown methodology for selecting practices which were early adopters to EHR and up to 5 years later adoption, introducing potential bias in physician practice type.

^gResults accounted patient years, not individual patients.