By Liz Szabo, USA TODAY Consumers are inundated with health advice from a host of sources: movie stars, TV docs and even the Internet. Thanks to the Internet, doctors say patients often come to their offices now toting pages and pages of medical information — with lots of questions about their health. But not all of the medical research on the Web is reliable, says Josephine Briggs, director of the National Center for Complementary and Alternative Medicine. CELEBS: Are they crossing the line on medical advice? Before following any medical advice from the Internet, a celebrity or any other source, Briggs and other experts suggest that people think critically and talk to their doctors. Jeffrey White of the National Cancer Institute says consumers also should watch out for these red flags: 1. A treatment is touted as a "cure," "miracle" or "breakthrough." If it seems too good to be true, it probably is. While many chronic diseases can be managed, there are relatively few true cures. People who promise dramatic recoveries may not be telling the whole truth. For example, someone may attribute her cancer recovery to dietary supplements, without mentioning she also had surgery and radiation. 2. There's no mention of side effects. All drugs — even natural ones — have side effects. Promoters should be honest about side effects, as well as possible interactions with other conventional or alternative therapies. Promoters should also be honest about whom a therapy is most likely to help and who should avoid the drug. 3. The promoter relies on anecdotes and personal testimonies. An anecdote may generate a hypothesis, which could lead to a clinical trial, but anecdotes, on their own, prove nothing. That's because anecdotes — including stories in which people made dramatic recoveries — can be misleading, because they don't tell you anything about the larger picture. Though 10 people may have done well taking an herb, hundreds of others may have gotten much sicker. 4. There are no results from published clinical trials. Randomized clinical trials are considered the gold standard of medical evidence, and they're the only way to prove cause and effect. To gain approval from the Food and Drug Administration, companies typically must perform large studies in which they randomly assign one group to follow one regimen, such as taking a new drug, while a comparison group does something else, such as taking an established therapy. This allows researchers to measure a drug's safety and effectiveness and also measure how much of a therapy's benefit, if any, is caused by a "placebo effect," in which people improve simply because they expect that they will. Consumers can look up published study results through sites such as clinicaltrials.gov and pubmed.gov.