BACKGROUND: Slow adaptation of new information by providers may result in suboptimal care. OBJECTIVE: To evaluate changes in prescriptions for combination hormone replacement therapy (HRT) after a multicomponent intervention to deliver new information to patients and providers. DESIGN: Quasi-experimental study with multiple baselines. SETTING: Veterans Affairs Tennessee Valley Healthcare System (VA-TVHS). PATIENTS: Female veterans age 50 to 79 years who had a prescription filled at the VA-TVHS for combination HRT between 1 January 2002 and 1 July 2002. MEASUREMENTS: Discontinuation of HRT. INTERVENTION: A 3-part intervention consisted of 1) notifying patients who were using combination HRT of the results of the Women's Health Initiative study (patient education component), 2) sending all providers an e-mail with the Women's Health Initiative study results (provider education component), and 3) placing an electronic alert in each eligible patient's chart (provider care component). The alert asked providers to reevaluate the need for combination HRT. The intervention was implemented at different VA-TVHS sites in a stepwise fashion to differentiate intervention effect from media effect. Study follow-up continued through 31 December 2002. RESULTS: The total rate of discontinuation of combination HRT was 70.3% in 2002. The proportion of discontinuation from time of media release until intervention was 23.3%. After initiation of the intervention, an additional 43% of the original cohort discontinued use of HRT; this percentage represents a 59% relative decrease in HRT use among patients. After adjustment for time, the discontinuation rate per day was 4.9 times higher after the multifacted intervention than after the media release (95% CI, 1.8 to 13.1). LIMITATIONS: A true control group is lacking. CONCLUSION: A multifaceted approach in an integrated health care system with standardized methods of communication is an effective way to implement patient-centered, effective, and timely care with changing medical knowledge.