Why (and how) you should encourage your patients’ search for health information on the Web
69% of Americans look for health information on the Internet. Don’t discourage your patients from this: Instead, empower them to identify reliable advice. Includes a patient handout to aid your efforts.
IN THIS ARTICLE
To obtain a PDF of this Patient Guide, click here .
Four tips for evaluating online content
Consider the source
The very first thing to consider is the domain name—e.g., “.gov,” “.org,” “.edu,” “.com,” or .anythingelse.
.gov sites are owned and maintained by the US government. From a medical standpoint, the .gov designation means that a site contains evidence-based information maintained by medical librarians that is written at a level that most people can read.
It’s a common misperception that the .org designation indicates a not-for-profit site that is therefore “looking out for the public’s best interest.” But anyone can purchase an .org domain. Even if a particular site really is administered by a not-for-profit organization, that status does not ensure that the content is of high quality.
.edu sites are affiliated with academic institutions.
.com and .anythingelse are, like .org, free for anyone to purchase.
Evidence-based medicine tells us that .gov sites are the most accurate for content—anything else is buyer beware (interestingly, .edu sites fared the worst in one study for accuracy).7,8 This doesn’t mean that sites that are not .gov have no value! On the contrary, non-governmental Web sites, blogs, and news articles are often the first place a search starts. That’s because current content, such as a new site, is more likely to feature prominently in a search engine response. But non-governmental sites do require an additional level of scrutiny. For example, in one study, only 55% of health information on news Web sites was medically accurate.8
Who is the author? Does she have financial ties or bias?
Think about who is running the Web site, why they are doing it, and what they are trying to achieve. Is the author a doctor, a health reporter, an advocate, or a drug company? The credentials of any physicians associated with the site should be listed as well as credentials and affiliations of authors, if they are non-physicians. Sites that list physician credentials tend to rate higher in accuracy of content.7
Talk with your patients about the importance of financial ties. The sponsor of the Web site should be listed (drug company, university, or a physician blogging without reimbursement). Bias and money go hand in hand, so be careful in evaluating whether the information provided favors the sponsor. If there are products for sale that are recommended by the medical content, bias is implied and, in my opinion, that site is not informational but commercial.
Bias can also be non-commercial, which is why authors of academic papers are not only supposed to report financial ties but also any real or perceived conflict of interest. This can be hard to discern at times, but the Web site should disclose why it exists. If the message is muddied by advertising and other commercial content, it’s best to just move along.
How current is the content?
We all know that new studies constantly reshape the way we practice (and sometimes guidelines from professional societies seem to change with the wind), and what is new quickly becomes out of date. The good and bad with online information is that it is always there. A permanent record is in many ways a good thing, but content from 2004 may not be applicable in 2011. This is a particular issue with news sites. They may report on a fascinating study in 2007, but if a retraction later appears or a new study refutes the findings, that information may not warrant an article on the Web. A good rule of thumb: Anything older than 2 years requires an additional level of scrutiny.
Be wary when the subject is complementary and alternative medicine
Online content related to complementary and alternative medicine (CAM) should be approached with a higher degree of caution. One study found that 25% of CAM sites presented information that could cause physical harm if acted upon, and almost all CAM sites omitted vital warnings, such as drug interactions, contraindications, and adverse reactions. The quality of CAM sites doesn’t improve even when they meet three or four of the JAMA benchmarks for information quality (see page 40).9 In one study of breast cancer sites, Web pages with CAM content were 15 times more likely to contain inaccurate content, compared with sites without CAM content.7
Instruments and tools for evaluating online content
Three tools are available to help patients and providers judge the quality of written online information:
Discern is a 16-question tool designed to assess the quality of online health information. You will find it at https://www.discern.org.uk/discern_instrument.php. I recommend that you mention this tool to patients, even if you are uncertain whether they will use it. Certainly, any patient wedded to what seems like questionable content from a specific Web site should be encouraged to evaluate the site using the Discern tool. In addition, if you have Web sites other than .gov sites that you like to share with patients, it might be wise to personally review them with Discern so that you can reaffirm that you are directing patients to reliable content rather than a quagmire of misinformation.10
