This 212-word news brief covers a lot of bases, but not enough. It quickly summarizes a new population-based, case-control study from Sweden which suggests that breast-feeding may be protective against rheumatoid arthritis. Unfortunately the news story reduces the study’s conclusion to a fact–breast feeding "reduces the risk of rheumatoid arthritis"–when, in fact, the researchers themselves say more research is needed to confirm their results. The story provides important hints about both the advantages and disadvantages of the study’s design—which are critical to understanding the limitations of the study’s conclusions. It mentions the study’s "large community-based approach," but that "it is difficult to separate the effect of breast-feeding from childbirth itself". Still, as the researchers themselves acknowledge, it is also possible that the study is too small to be conclusive or that they simply overlooked some other risk factor(s) that could account for the reduction in RA. Moreover, the news story fails to ask whether the reduction in RA risk is really meaningful. How many lives are we talking about? It would appear that the number protected by breast-feeding is extremely small—reductions of mere hundredths of one percent—almost surely too small to accurately measure in a study of this type, or any other (see Quantification of Benefits…" below). If the news story had sought out an independent source, it might have found someone to offer a different perspective on the lead author’s belief that even a month or two of breast–feeding "makes a difference."
Not mentioned but also not terribly important.
The news brief says women who breast-feed reduce their risk of rheumatoid arthritis by 25% to 50%. But ideally, readers would also learn the absolute numbers on which these estimates of relative benefits are based. However, the published study also failed to report the reduction in absolute numbers. What’s a reporter to do? Run the numbers on a calculator? Ask the researchers for more information? If the estimated prevalence of RA is 72/100,000, a 25% reduction might be somewhere in the neighborhood of 18/1000,000 or about two-hundredths of one percent (0.02%), and a 50% reduction might be an absolute reduction of 0.04%–a small number any way you look at it. At the least, readers should be made aware that the true meaning of these numbers is uncertain.
The study cites the health advantages of breast-feeding. Are there no potential harms or disadvantages? For example, there is some research to suggest that pregnancy and breast-feeding may increase flares of RA.
The news brief describes the methods of the new research in enough detail for readers to accurately infer that it was a case-control study, with results adjusted for factors that might affect the incidence of RA. The story mentions some of the pros and cons of the study’s design as well–the advantages of a "large community-based approach" and the fact that "it is difficult to separate the effect of breast-feeding from childbirth itself." However, the story does not mention other limitations inherent in the study’s design that could also undermine the results. Is it possible that the study simply overlooked some other factor(s) that might account for the reduction in RA incidence among Swedish women who breast-feed? Can women accurately recall today the breast-feeding habits of children born decades earlier? Was the study large enough? The new research is suggestive, not definitive—but the news story does not make this sufficiently clear. (See also "Quantification of Benefits…" below.)
The news story does not hype or oversell rheumatoid arthritis, an often-debilitating but relatively uncommon disease. On the other hand, it also does not report the prevalence or incidence of rheumatoid arthritis, which would be helpful. Estimates of annual incidence vary; in the UK the estimated incidence is 36/100,000 of women of all ages, while in the new Swedish study the estimated incidence was double that, or 72/100,000.
This story cites a single source, the lead researcher of the study summarized in the news brief. No independent source is cited.
The story does not mention whether other factors (e.g. behavioral, environmental, genetic) might also reduce the risk of RA. For example, one of the purposes of the Swedish study was to see whether oral contraceptives—containing hormones that rise during pregnancy—protect women against rheumatoid arthritis. The study concluded that they do not.
Most readers will be familiar with the "availability" of breast-feeding.
Breast-feeding has been around for millennia, and most readers know this.
There is no obvious use of text from a news release.