Posted on 02/12/2014 at 11:48:23 AM by Student BloggerBy Lindsey Smith, MPH
Imagine you're browsing your morning edition of NPR, maybe checking out some of the hilarious and alarming stories circulating about the Sochi Olympics, when a flashy headline catches your eye. “Cash or Credit: How Kids Pay for School Lunch Matters for Health.” Oh really? You might think. How does that work, exactly?
Upon finding the original study, published in the journal Obesity in 2013, you discover that the study was not a randomized-controlled trial (RCT) testing whether using cash or credit is associated with some health outcome (obesity, perhaps, or maybe insulin resistance). Rather, it is a cross-sectional study examining whether attendance at a school with different methods of payment for school lunches is associated with certain diet behaviors, like eating higher total energy or purchasing fresh fruit or dessert. The NPR article tells you that students at schools with debit-only systems have increased total energy intake and lower nutritional quality compared to schools with debit-or-cash systems, suggesting that purchasing lunch with a debit card somehow causes students to make less healthy choices.
A closer look reveals a more complex picture. As it turns out, the study itself did not examine which method the students themselves used, only what type of school they attended. Could it be that the observed differences between payment systems were actually due to some other characteristics of the school that are associated with eating more healthfully? As it turns out, Whites were more likely to attend debit-only schools compared to Blacks, while low-income kids were more likely to attend cash-or-credit schools than higher income kids. Unfortunately, the authors did not control for these differences in their analyses, making it even more difficult to draw conclusions from this research. In short, this study demonstrates school-level associations between school lunch payment systems and dietary intake, with no evidence to suggest causation at the individual level. So what's with the splashy headline, then?
Selvaraj et al. set out to answer this question in a recent study published in PLOSOne, entitled “Media Coverage of Medical Journals” Do the Best Articles Make the News?” The authors argue that poor health care journalism is the result of not only the quality of the coverage, but the type of studies covered. Moreover, when media outlets choose to cover studies of poorer quality, they forego coverage of rigorously conducted studies, which could have important implications for public health. To investigate whether the lay media systematically cover studies of weaker design, the authors compared study characteristics of 75 clinically-oriented studies that received coverage by top newspapers against 75 studies published in the top five medical journals over a similar time range.
The authors found that studies covered by newspapers were significantly less likely to be RCTs, but more likely to be observational studies. Looking more closely at the study designs, they found that observational studies covered by newspapers were more likely to have smaller sample sizes and be cross-sectional, but there were no differences between newspapers and medical journals regarding RCTs. So yes, it turns out that, according to this study, lay media selectively reports observational studies, which cannot determine causality, but often assign them this causal role, ascribing them with an importance and potentially a larger health impact than the study design warrants.
Of course, any study design has limitations and nuances that complicate interpretation, but lay audiences cannot be expected to understand all of these caveats, especially not while casually perusing the news. As a result of such selective reporting and headline inflation, the general populace may make decisions about nutrition and health that are not supported by the science.
However, the responsibility for communicating the correct research findings does not lay entirely with the journalist. To improve the likelihood and quality of lay publication, Selvaraj et al suggests that scientists hold more and better press conferences. More research is also needed to understand why some studies are promoted via press release while others languish, and how to incorporate expert consultation to help identify which studies are worthy of coverage.
In the meantime: read with caution.
Just DR, Wansink B. School Lunch Debit Card Payment Systems are Associated with Lower Nutrition and Higher Calories. Obesity 2014;22:24-26.
Selvaraj S, Borkar DS, Prasad V. Media Coverage of Medical Journals: Do the Best Articles Make the News? PLOS One 2014.