A parent’s guide to health information on the Internet

The Internet can be a rich source of information on child and youth health, but it can also be a source of confusion and contradiction. How do you know whether the information you’ve found is reliable?

High-quality information should be objective and based on scientific evidence where available. When advice can’t be backed up by solid evidence, the next best option is the opinion of experts in that field of health care.

When Web sites are trying to sell or promote a product or opinion, they are biased. Often, there is no scientific basis to back up the claims they make.

When you’re looking for health information, the two most important things to consider are the following:

• Who is hosting the Web site?

• Is the information high quality?

Who is hosting the Web site?

Many Web sites are designed to sell products or promote a particular opinion. If the person or organization hosting a Web site will make money if their advice is followed, then they are in a conflict of interest. Sometimes a conflict of interest is obvious, but other times it is harder to tell. Here are some questions to ask:

• Is the site trying to sell a product or service?

• Is the site hosted by a for-profit organization? Look for an ‘About us’ or ‘Contact us’ link, where you can find out more about the host. Even if some sites aren’t selling products, they may still be sponsored by companies that will benefit from sales if the advice on the site is followed. If a site is anonymous, then you should disregard it.

• Is the site sponsored by a for-profit organization? Check to see whether the site sponsor is providing an ‘unrestricted grant’. This usually means the site sponsor is not involved in the content.

• Are there advertisements or ‘pop-ups’ on the site? Selling ad space can make a site quite profitable, and controversial opinions about health issues can increase traffic on a site. However, some good sites also sell advertisements to cover their costs and are not in a conflict of interest.

• Is the site asking for your personal information? Your e-mail address may be used later for advertising or promotion.

• Are you being asked to fill in a questionnaire? Sometimes sites try to get information about what you like, or what products you use, and this information could be used for marketing purposes.

• Is the site hosted by a widely recognized authority on the subject? Examples include national organizations that government uses for guidance. Watch out for special interest groups, which may publish information that is supported only by a small number of practitioners, and may not be based on evidence.

• Is the information ‘peer reviewed’? That means that it has been reviewed and approved by other experts in that field. The site should specify the type of peer review (for example, a board or committee of experts that is not in a conflict of interest). Web sites that are not peer reviewed may contain unproven claims that seem too good to be true. Unfortunately, they often are.

A conflict of interest may exist even when the goal doesn’t involve making money. How can you tell whether a site has a hidden agenda? It can be difficult. It is also difficult to distinguish between good health information Web sites from sites that publish honest mistakes.

Is the information high quality?

High-quality information is based on up-to-date evidence that comes from proper research. Health information can be biased not only by a conflict of interest, but also by errors in how the research was done. Unless you have special training in research methods, it may be hard to tell whether the health information you’ve found is based on proper research. Here are some questions to help you judge the information from a scientific point of view:

• Is the information recent, and does the Web site indicate when it was posted?

• Is the advice based on opinion or is there true evidence to support it? Even expert opinion is still just a point of view. The opinion of recognized experts may be your best option, but only when good evidence is unavailable. Sometimes evidence isn’t available when certain research questions are too difficult to study or the research simply hasn’t been done yet. When clear conclusions can be drawn from solid evidence, opinion that differs from it – even if it comes from recognized experts – should be disregarded.

• Does the site describe the kind of study on which the advice is based? Is the advice based on evidence taken from research data that are collected retrospectively (from old, often incomplete, records or memories, which are usually less reliable) or from data that are collected prospectively (on a go-forward basis, which is more reliable but often more difficult to study)?

• Experimental studies can be set up to reduce personal bias and bias that comes from systematic error. In observational studies, the investigators do not get involved; they just ‘see what happens’. Observational studies are more likely to contain some unrecognized bias. The strength of the evidence depends on whether it is observational (weaker evidence, and more subject to bias) or experimental (more convincing but often more difficult to study). Examples of studies with observational information include:

  • Testimonials or anecdotes (stories): “From our experience…”

  • Case reports (studies on a single patient).

  • Case series (studies on just a few patients).

  • Case control studies (‘observational’ studies on patients, where patients with the condition of interest are compared with individuals who do not have the condition of interest).

  Using statistical methods to ‘adjust’ for possible bias (of which the investigators are aware) can reduce the bias in observational studies after the data are collected. However, although experimental study designs usually have less bias to begin with, they can be biased too. Examples of experimental studies that are designed to reduce bias are (in order of studies that have the most to the least potential for bias):

  • clinical trials;

  • controlled clinical trials;

  • randomized, controlled clinical trials; and

  • double-blind, randomized, controlled clinical trials (this study design produces the strongest evidence). In these experiments, a treatment is compared with a ‘control’ (it may be a fake or ‘placebo’ treatment, or a different kind of treatment). Neither the parent/child nor the doctor knows whether the child received the treatment until the end of the study. The patients are assigned to either a treatment or control group at random (like tossing a coin). Randomly assigning patients to each group reduces the chance that the patients with more severe disease are put into one treatment group or the other.

This guide may help you rule out some Web sites as sources of misinformation, but it won’t be enough to help you choose the best sources of quality information to make important health decisions for your family. Be sure to discuss information you find with your child’s doctor.

For more information

• The Health on the Net Foundation accredits medical Web sites that follow a voluntary code of conduct, and can be found at <www.hon.ch>.

• Guiding parents in their search on the Internet for high-quality health information, a Canadian Paediatric Society practice point, developed by the Community Paediatrics Committee, is available at <www.cps.ca>.