OBJECTIVES: The objective was to summarize the impact of pharmacists’ interventions on patient outcomes as reported in studies utilizing a randomized controlled trial (RCT) design.  METHODS: A comprehensive literature search was conducted utilizing PubMed and International Pharmaceutical Abstracts for the years 1979-2009. Studies were included if they evaluated pharmacist-provided interventions, utilized RCT designs with control groups, and were conducted in the United States. Studies were summarized by 1) publication year; 2) study setting; 3) disease/health condition; and 4) type of intervention provided and whether performed by a pharmacist alone or a pharmacist with other health care professionals, 5) primary outcome variables and 6) study findings.  Patient outcome results were categorized as ‘positive’, if they found a statistically significant improvement in the patient outcomes when compared to the control group, or ‘no difference’ if no significant difference was found.     RESULTS: Of the 552 citations reviewed, 100 studies met the inclusion criteria. The numbers of pharmacists’ intervention studies conducted using RCT designs increased from 1979 to 2009, with 64% conducted after 2000. The majority of the studies were conducted in clinic setting (65%) followed by community pharmacies (19%), hospitals (12%), home-care (2%), and multiple settings (2%). Chronic conditions were studied in 56% of the studies. Studies of interventions delivered by pharmacists alone comprised 64%, while 36% involved pharmacists working with other health care providers. Overall, ’positive’ patient outcomes were demonstrated in 73% of the studies. When interventions were delivered by pharmacists working with other health care providers, however, 97.2% had positive outcomes. CONCLUSIONS: ‘Positive’ patient outcomes were demonstrated in the majority of the studies, supporting increased pharmacists’ roles in the health care system to improve patient outcomes.  When working alongside other health care providers, the positive impact was most likely.