Introduction
Consumers looking for health information on the Internet have access to tens of thousands of health-related Web sites and literally millions of Web pages of health-related materials. Because almost anyone can make a Web site available through the Internet, the breadth of information that can be found is vast. On health topics, consumers can readily be linked to scientific information about various health problems and treatments, advertisements for physician and hospital services and pharmaceuticals, advocacy organizations, accounts of personal experiences, and opportunities for people with similar interests to interact online. Studies also indicate that consumers are paying attention to what they find; more than 70% say online health information has influenced a treatment decision (Fox and Raine, 2000).
The rise of consumerism in health care has resulted in increased demand for information. Doctors are interested in patient education because it enhances the ability of patients to participate in their care and may also increase compliance with treatment regimens. However, the process of finding high-quality, appropriate health information can be an arduous task. Several studies of single medical conditions have suggested deficiencies in the quality of Web-based health information (Sandvik, 1999).
The ways that this enormous amount of material will change health care delivery is structured and provided remains to be seen. Optimists predict that ready access to the latest scientific information will allow consumers to make more informed decisions about their health care. Because the information is free, it has been suggested that some disparities in access might be eliminated (United States Bureau of Commerce, 1999). Others worry that the Internet contains incomplete and misleading information, which may directly harm patients by misdirecting a treatment decision (Ostrom, 1999; Pear, 2000). Another area of recent concern is that of direct access to pharmaceuticals without visiting a physician. There is adequate evidence on the Internet for all of these positions (Armstrong et al, 1999).
Over the past year, policy makers have become increasingly concerned with the problem of how to ensure that the information consumers find on the Internet is of high (or at least acceptable) quality (Landro, 2000; Baur and Deering, 2000). Several individuals and organizations have proposed and developed guidelines to evaluate health-related Web site content (e.g., HON Code, American Medical Association, Internet HealthCare Coalition, Hi-Ethics, MedCertain), but these guidelines have not been systematically applied to a broad set of Web pages and conditions. Furthermore, little scientific research has been conducted to empirically assess what a consumer might find while searching on the Internet.
While accessible and high-quality health information on the Internet is important for English speakers, it could be even more useful for Spanish speakers, who often encounter greater barriers to traditional sources of medical care and information. Preliminary data from the 2000 US Census indicate that Hispanics are now the largest minority population in the United States.
In response to these concerns about the variation in quality of health information on the Internet, RAND collaborated with the California HealthCare Foundation to design and implement the Evaluation of English and Spanish Health Information on the Internet Project. This project represents one of the most comprehensive and systematic efforts ever undertaken to evaluate health information available on the Internet. We believe that it is the first project to simultaneously evaluate English-language and Spanish-language search engines and Web sites. In designing this study, we addressed three questions:
This study focused on four medical conditions (breast cancer, childhood asthma, depression and obesity) in two languages (English and Spanish). These are conditions that affect diverse populations and are associated with an increased likelihood of early death and disability. Thus, they represent the kinds of health care concerns that would motivate consumers to look for information on the Web.
Chapters 2 through 4 follow a standard format. We begin by explaining the purpose of this investigation. We then provide a detailed description of the methods employed. In many cases, we had to develop new methods or adapt existing methods to study the Internet. These methods may be useful to others who are interested in evaluating other search engines, other health topics, or Web sites other than those selected for this study. Next we provide the results of our work. We end each chapter with a discussion of the findings.
In Chapter 2 we present our assessment of the performance of 14 frequently-visited search engines (10 English-language and 4 Spanish-language engines). In Chapter 3 we report on our evaluation of the comprehensiveness, accuracy, and currency of information presented on general health and condition-specific Web sites. In Chapter 4 we report the reading grade level required to understand information presented on selected Web sites. Finally, in Chapter 5 we summarize the results and make recommendations.