Here’s the study.(via reddit) It states that using the statistics available from three very different pools of Muslims, Ramadan fasting can be shown to have clear negative effects on the health of a child at the statistical level. This is, as one can imagine, an inherently incendiary proposition. It ties together a nuanced scientific point with an easily-manipulated set of political implications and eyeball fodder, and we all know where those tend to end up. But it’s worth digging in to the details in order to be prepared should this hit the headlines of the punditry circus. Some background: while pregnant women apparently have flexibility in determining how they will handle Ramadan, some apparently choose to practice different forms of fasting. I have no idea how this is handled in more traditional societies. As far as the study goes, page ii/section A.1.1. addresses the problem of the scarcity of data available and the effect of using a “repay fasting after child delivery” structure, namely that people don’t like fasting on their own and they don’t like being left out of community rituals.
Second, because it will come up near instantaneously, these are economists surveying medical data, and that’s why it comes from that source which oversees the Michigan-area Muslim minority.
The conclusions (p29) suggest that the most serious impacts of fasting on a pregnancy happens during the first month, which is also when a woman is least likely to be aware of her pregnancy. They also draw a connection to how an estimated 40% of US women of childbearing age are attempting to lose weight. In general, the results might be best described (for the US) as providing a concentrated example that should suggest that it’s in everyone’s interest to screen frequently for pregnancy before and during any sort of dietary reduction program at intervals smaller than 1 month. In a way, it’s one of those public health issues that is subtle enough that while Ramadan provides an easy indicator for study, the effect of the information derived is of universal utility assuming it is handled with a certain sensitivity to the source of the data. At the very least, it merits further study from medically trained analysts in order to decide what the cost-benefit really looks like and what, if anything, should be done.
Nevertheless, the off-the-cuff description of this study seems so perfectly designed to generate hysteria that I thought it would be useful to toss out some ideas on it before it it ends up summarized into madness.