– Acceptable research sources include: researchers, official associations, university/medical centres (such as the Mayo Clinic), government research, full text of peer-reviewed studies, or university-based press releases.

– Secondary sources: health news sites (WebMD, Science Daily, etc.) are not acceptable on their own.

– Never use Wikipedia or blogs as primary sources.

– All research cited as recent should be within the last five years.

– If the full text of a study is not online, the writer must contact the lead researcher.

– Take the time to place research in context. Don’t consult a single study; make sure it exists within a framework of supported research.

– Know the difference between population studies, animal studies and human studies.

– Population studies may indicate types of behaviours that are associated with a particular health benefit (for instance, people who drink green tea are less likely to develop heart disease) but rarely with a causal relationship (green tea does not prevent heart disease) because it is just as probable that other factors are involved (people who drink green tea may be less likely to consume meat-lovers pizzas more than once a week).

– For animal studies, please note the dosage of a particular drug or food – it is often significantly higher than is safe or practical for human consumption. Please do not extrapolate animal studies to human health outcomes.

– For human studies, please be careful to note how that group is selected or limited, and, whenever possible, use studies that use normal, healthy, adult women (what helps weight loss among post-menopausal obese women with Type-1 diabetes may not pertain as directly to an average, healthy 45-year-old woman.

To read more on health-reporting, check out my article “The media diet” in the winter 2013 issue of the Ryerson Review of Journalism, which will be available throughout Canada on most major newsstands just before or shortly after Christmas.