Health information technology (IT), such as computerized physician order entry and electronic health records, has potential to improve the quality of health care. But the returns from widespread adoption of such technologies remain uncertain. We measured changes in the quality of care following adoption of electronic health records among a national sample of U.S. hospitals from 2004 to 2007. The use of computerized physician order entry and electronic health records resulted in significant improvements in two quality measures, with larger effects in academic than nonacademic hospitals. We conclude that achieving substantive benefits from national implementation of health IT may be a lengthy process. Policies to improve health IT's efficacy in nonacademic hospitals might be more beneficial than adoption subsidies.