An article published in Pediatrics, Respiratory tract diseases during the first year of life: effect of dog and cat contacts "is getting a lot of attention in the news, but more than misses what most of these stories generally Miss: you cannot prove cause and effect from an observational study. And there are major limitations on research based on people keeping diaries and answer questionnaires.
The study concludes:
".. dog contacts may have a protective effect on the respiratory tract infections during the first year of life. Our findings support the theory that, during the first year of life, contacts are important, possibly leading to greater resistance to infectious respiratory diseases during childhood "
The Wall Street Journal reported:
"While the study controlled just under 400 babies, the researchers said that the results were statistically significant because it relied on questionnaires completed by parents. weekly" (Our reaction: huh? This sentence makes no sense. Statistical significance is not determined by filling out questionnaires from parents.)
WebMD reported:
"It is not clear why live with a dog makes a difference." (Our reaction: it is not clear that living with a dog makes a difference. Making a difference means you have already proved the cause and effect and this study did not do it.)
HealthDay reported:
"Exhibition of cats also showed a protective effect, but was not as strong as the effect of exposure of the dog". (Our reaction: no protective effect was created in this observational study.)
None of these news organizations earlier and any of the following mentioned the limitations of observational studies and that it is not possible to establish a causal relationship. Reuters Health nor the Los Angeles Times, or CNN. (90 minutes later Addendum: Nor CBS. Or TIME.com. Neither the Toronto Star. )
From our early morning sweep, just a news story of my HealthNewsDaily the MSNBC.com included what we were looking for:
"The relationship between pets and less infections held even when the researchers took into account factors known to affect infection rates of infants, how to breastfeed and number of brothers. Still, the researchers acknowledged that it could not account for all factors and noted that they found a correlation, not a relationship of cause and effect. "
That wasn't so hard, was it?
Our usual reminder: journalists and consumers must read our booklet, "the language fits the evidence? Association Versus causation. "
0 comments:
Post a Comment