BackgroundThe extensive availability of online health information offers the public opportunities to become independently informed about their care, but what affects the successful retrieval and understanding of accurate and detailed information? We have limited knowledge about the ways individuals use the Internet and the personal characteristics that affect online health literacy.ObjectiveThis study examined the extent to which age and cognitive style predicted success in searching for online health information, controlling for differences in education, daily Internet use, and general health literacy.MethodsThe Online Health Study (OHS) was conducted at Johns Hopkins School of Public Health and Stanford University School of Medicine from April 2009 to June 2010. The OHS was designed to explore the factors associated with success in obtaining health information across different age groups. A total of 346 men and women aged 35 years and older of diverse racial and ethnic backgrounds participated in the study. Participants were evaluated for success in searching online for answers to health-related tasks/questions on nutrition, cancer, alternative medicine, vaccinations, medical equipment, and genetic testing.ResultsCognitive style, in terms of context sensitivity, was associated with less success in obtaining online health information, with tasks involving visual judgment most affected. In addition, better health literacy was positively associated with overall success in online health seeking, specifically for tasks requiring prior health knowledge. The oldest searchers were disadvantaged even after controlling for education, Internet use, general health literacy, and cognitive style, especially when spatial tasks such as mapping were involved.ConclusionsThe increasing availability of online health information provides opportunities to improve patient education and knowledge, but effective use of these resources depends on online health literacy. Greater support for those who are in the oldest cohorts and for design of interfaces that support users with different cognitive styles may be required in an age of shared medical decision making.