Even so, definitive conclusions can't be made from such a comparison due to a number of
confounding differences between the studies.
Not exact matches
It came up at christmas dinner; at one point, the topic of breastfeeding came up, and my spouse's uncle started to say «I know breastfeeding reduces the risk of allergies later» and, expecting him to move on to obesity and IQ and whatever else he was about to say, I just said «No, that's not true», and started trying to explain the
difference between correlation and causation, and the difficulties with prospective
studies and
confounding factors, etc..
I know it's really difficult to correct for
confounding factors in a
study like that but if we're really talking about «the future of mankind» shouldn't we be seeing some really profoundly alarming
differences between the two?
Although breastfeeding in contemporary, industrialized nations is associated with higher social class, IQ
differences between breastfed children and children not fed breast milk remain significant in most observational
studies even after adjustments for class - related
confounding factors (16, 17).
However there was some evidence that more pronounced treatment effects were associated with
studies at higher risk of bias; this could potentially
confound any
differences between subgroups.
«We did not find any
difference between children who received general anesthesia and spinal anesthesia, suggesting that the previous human
studies may have been affected by
confounding factors.»
Writing in a linked Comment, Dr Alexander C McFarlane, University of Adelaide, Australia, notes that «a potential
confounding issue with this
study design is the fact that by filling in the self - report measures, both groups had the opportunity to reflect on their obvious levels of symptomatic distress... which might have minimised the ability to find
differences between groups.»
In discussing the limited evidence for the «probable» link
between red meat and colorectal cancer, the WHO itself concedes that it is not possible to rule out other explanations (which it helpfully describes as «chance, bias or
confounding»).2 Harcombe agrees, arguing that even when
studies strive to adjust statistically for baseline
differences in relevant factors such as socioeconomic status, body mass index, physical activity, smoking status and diabetes, it is impossible to grapple fully with all the factors that differentiate «the couch potato» from «the paleo buff» (her ideal), or to take into account the «chasm» that separates fresh and traditionally preserved meats from modern manufactured meat products.9
First, although the within - person design of these
studies helps rule out the influence of
between - person
differences, and although we controlled for important within - person
confounds, it remains possible that other
confounding factors not controlled in our analyses account for the interactive effects of HC discontinuation.
Though the general trend of the data from this
study supports the Cinderella effect, Anderson and colleagues note that the observed
differences between parental investment in genetic children and stepchildren might be slightly reduced by a few
confounding factors.
This
study includes a rigorous active control condition meaning any
differences between conditions can be confidently ascribed to improvements in executive functioning rather than other
confounding factors.