A trial that randomized people with Type 2 diabetes to drink one glass of either wine or water daily found that there might be some benefits to drinking the wine, according to this story. How much? Some. How much is that? The story doesn’t pin it down. The story is devoid of quotes and specifics, giving readers a vague sense of benefits, while missing key limitations in the study or warnings about risks that the researchers clearly pointed out in their journal article.
As our criteria emphasize, news stories about research results should be specific, giving readers hard numbers and relevant context with which to gauge the value of the thing being studied, in this case wine with dinner for people with Type 2 diabetes. Vague descriptions of “some” benefit are not enough, especially when those potential benefits appear to be quite modest.
The story does not report the cost of wine or mineral water, but these drinks are so commonplace that it seems reasonable to assume that readers already know what they cost. So we’ll rate this Not Applicable, but we’d note that a typical bottle of wine is 750 ml, which is 5 servings of 150 ml (about 5 oz) that were ‘prescribed’ in the trial. Although the cost of a moderately priced bottle (say $10) every 5 days is not huge, this could be a burden for some people on a fixed income. A better choice for them might be more fresh vegetables.
The story relies on vague descriptions of benefits. The only specific benefit mentioned is a 10 percent increase in HDL (“good”) cholesterol, and even in this instance readers are not given absolute numbers or any information about the importance of such a change. (This 2 mg/dL increase in HDL cholesterol among the red wine drinkers is very modest, and it is not clear that this is clinically important.) Otherwise, the story merely makes broad references to improved cardiac health, improved triglyceride levels, better sleep, fewer symptoms of metabolic syndrome and so on without quantifying the differences or explaining whether they were meaningful to the health of the individuals in the study.
In their journal article, the researchers detailed how modest and sometimes murky the results were. The story presents the results in a far bolder and simpler manner, giving readers an exaggerated sense of the clarity and power of the observed effects. What we need to know (and don’t get from the story) is whether these people will have a decrease in the important health outcomes of heart attacks and strokes in the future as a result of the wine drinking.
Although the researchers did not report adverse events in their journal article, they described at length how they excluded several categories of participants, including smokers, pregnant women, those with a family history of addiction, women with close relatives who had breast cancer, anyone under 40 and others, in order to minimize the potential risks of drinking alcohol. Other risks of alcohol include less common events such as falls, which may not be apparent in a smaller study over only 2 years. In addition, the public tends to generalize, and the take away from this story for some readers might be that red wine is good for everyone with diabetes. In fact, this study was only among well controlled type 2 diabetics who were either on oral agents or insulin no more than once daily. The researchers explicitly warned about the potential risks of recommending that people with diabetes drink wine. That concern about harming people who might be vulnerable to adverse effects of alcohol was left out of the story.
This ruling is a close call.
The story tells readers about some of the key strengths of this study, that it is the first long-term randomized controlled trial of alcohol of its type, that it compared white and red wine in order to look for possible effects of components other than alcohol, and that a genetic marker of alcohol metabolism was measured, which revealed differences in how people responded according to which gene marker they have.
But the story fails to report important limitations of the study. It doesn’t tell readers that it involved fewer than 250 participants. The story reports the trial was conducted by researchers in Israel, Sweden, Germany and the United States, giving it an international feel, without telling readers that all the participants came from two centers in Israel.
Lastly, the story should have pointed out that cholesterol (and other markers studied such as blood pressure and waist circumference) are markers or surrogates for more important outcomes that matter to patients. What we really need to know is risk of heart attack, stroke, and death down the line.
The story is clear that this study is relevant to people who have Type 2 diabetes.
There are no quotes or other indications that independent sources were consulted. An outside perspective, particularly from someone who could point out that the benefits were very modest clinically, would have been valuable.
There is no discussion of how the effects seen in this study compare with other methods of improving cardiac health, managing cholesterol or sleeping better, including lifestyle or medications.
Moderate exercise, for example, also raises HDL levels, lowers blood pressure, and reduces waist circumference. And yes, it also improves sleep and mood, and reduces fall risk.
It is obvious that red and white wine are readily available to most readers.
The story notes that this study is based on randomizing people to either drink wine or mineral water and then following them for two years, in contrast to most studies in this area, which are generally based on observational data or small, short-term trials.
The lack of any direct quotes and the absence of any comments from independent sources make it difficult to determine whether or not this story relies largely on a news release. But since we didn’t see any obvious similarities between the story and two news releases that were issued about the findings (see here and here), we’ll give the benefit of the doubt on the rating.