Another overly optimistic account of findings from an observational study.
In this case, higher intake of soy foods was linked to better odds for survival after a lung cancer diagnosis in Chinese women. It’s potentially an important finding, but the story didn’t include the cautious language or discussion of limitations that we feel are mandatory when reporting on these kinds of studies. No attempt at critical analysis is made, and all findings are taken as truth.
Lung cancer is an aggressive disease with a low five-year survival rate. Any improvement here would be a step forward for patients and their families. However, the story and the research it cites do not provide compelling evidence that lung cancer patients should adopt a high soy diet.
The cost of soy foods is not in question.
The story is careful to describe the amount of soy food you’d need to eat on a daily basis to match the intake of the high soy consumers group — important info to include. But we were disappointed that the story used relative risk figures to describe the benefits of increased soy intake on cancer survival. It says that low soy consumers were “1.8 times as likely to die” compared with the average and that “Those who ate the most were about 11 percent less likely to die.” It would have been more informative and accurate to tell readers the absolute percentage of subjects who had died at follow-up in the low, average, and high soy consumption groups.
The story didn’t address harms, but we’re not sure that there are any harms worth mentioning when it comes to soy foods eaten in moderate amounts. Hypothetical concerns that soy may promote estrogen-sensitive cancers have not been confirmed in large studies. And the story correctly advises readers not to take soy supplements, which deliver a higher dose of soy isoflavones that might potentially increase cancer risk. We’ll rule it not applicable.
The story oversells what this study is capable of telling us. This story does not explain that the small cohort study based in China is very weak evidence for American women with lung cancer to add soy to the diet for the purpose of improving survival. This was an observational study designed to identify associations between soy intake and cancer survival. The limitations of a questionnaire-based study – with only two questionnaires filled out two years apart – deserve to be mentioned. The study was not capable of showing that soy “may … help people survive at least some forms of cancer better,” which is how the story frames the discussion in the lead sentence. There was also no mention of the many different factors that might have confounded the results. For example, women who eat a lot of soy might have other healthful habits that improve their odds of surviving longer.
This was a tricky call. The story calls lung cancer “the No. 1 cancer killer across the world,” and gives various lung cancer statistics describing a huge toll in the United States. The story qualifies these statements by noting that the Chinese women in this study were overwhelmingly non-smokers, and that lung cancer in non-smokers is probably a different disease than in smokers. But it never completely closes the loop on what this all means for women in the USA and other Western countries who are trying to make sense of these findings. Smoking accounts for some 87% of lung cancer deaths in the USA, according to the American Cancer Society, and the rate of lung cancer in nonsmoking women is much lower in the USA than it is in China. The upshot is that the findings probably have much less relevance for women in USA than they do for Chinese women. The story should have stated this more clearly.
Barely satisfactory – only because the story quotes a lung cancer expert who was not involved with the study. But that expert did not provide any useful criticism of this minimally important study, including the fact that all the findings could have been related to other non-dietary factors.
Smoking cessation would be far more effective than increased soy consumption for reducing lung cancer rates in American women. The story was not clear about this fact.
The availability of soy foods is not in question.
The story mentions previous research showing an association between intake of soy and rates of new lung cancers.
The story does not seem to have relied inappropriately on this press release. Plus there was an independent expert comment.