The “naughty” food or drink with a hidden benefit to health is a media perennial. Such stories have a deliciously unexpected quality: we all like to be told that our indulgences might not be unmitigated dietary disasters.

The usual suspects are red wine and chocolate, but two new ones have been added this week. White wine, as well as red, can protect the heart, according to reports. And another set of headlines told us that “jam can curb cancer”, as The Telegraph had it.

Both claims were rooted in respectable scientific studies. But as is often the case with such material, the findings were then over-interpreted and stripped of nuance to create a health message that is premature and misleading.

Take white wine first. It contains antioxidants called polyphenols, which are thought to contribute to red wine's moderate cardio- protective effect. Dipak Das, of the University of Connecticut, wondered whether white wine might thus share these benefits, and tested his hypothesis on rats.

A group of rodents were fed white wine as part of their diets, while others received water or grain alcohol. Some of these rats subsequently had heart attacks and those given the wine suffered less cardiac damage than those in the control groups.

This is an interesting effect, for sure, but not quite sufficient to justify the claims that were made on the back of it. Marie Claire's website, for example, declared: “We all know a glass of red wine a day is good for your heart, but did you know white wine may have just as many health benefits?”

It might, but then again it might not. A single study of a few rats certainly cannot tell us. While the benefits of moderate red wine consumption are attested by research into human populations, Dr Das's experiment has looked only at animals. It offers no reason for anyone to change their drinking habits.

The jam story is an even worse example of journalistic extrapolation. The study, from the Institute of Food Research, investigated a carbohydrate found in fruit and vegetables called pectin. Under the right conditions, this breaks down into other chemicals, which can impair the activity of a protein that plays a part in the spread of cancer. The scientists examined how one of these pectin derivatives binds to the cancer protein at a molecular level. Their tests did not even involve living cells, let alone animals or people.

The results illustrated a potential mechanism by which diets rich in fruit and vegetables might protect, as they do, against cancer. But they had nothing to do with jam. That angle emerged purely because the institute's press release mentioned that pectin is well-known as a gelling component of jam.

What followed was an absurd logical fallacy. Jam does contain pectin. Pectin does contain a chemical that inhibits a cancer protein in the laboratory. But it does not follow that jam must therefore “curb cancer”. Many biological effects that can be observed in petri dishes have no real impact in the human body. The hypothesis that jam prevents cancer is not even not proven; it has not been tested.

As the NHS Choices website puts it: “From reading some of the news coverage, one may get the impression that eating jams and jellies would have some anti-cancer effect. However, this is very preliminary research and such a conclusion would be premature.”

Preliminary research, of course, is important to understanding good health. It is a starting point for detecting and explaining effects that might be helpful or harmful. Great care must always be taken, however, not to over-interpret its results.

It does not matter much to people whether white wine protects rats against cardiac damage, or pectin derivatives bind to cancer proteins in-vitro. What counts is whether they stop people dying of heart attacks or tumours, and on such matters, studies like these have nothing to say.

Mark Henderson is the science editor of The Times

Explore Health

• Expert Advice

• Health Features

• Mental Health

• Alternative Medicine

• Child Health

• Health Club