[doukdouk]

In this case, the deworming treatment was allocated randomly (see the paper [0]). Giving the deworming treatment to children make them less poor than those who did not get the treatment, and there should be no baseline differences on average between the two group of children since the allocation was random.

Is going from "deworming children make less poor" to "the worms cause poor outcomes" such a stretch?

[0] http://emiguel.econ.berkeley.edu/research/twenty-year-econom...

[peter_retief]

There are so many factors here, poor health can exacerbate worm infestations and healthy individuals with a good diet can live with parasites for many years with no adverse symptoms. Then there are the visits to clinics that give default deworming pills as part of any consultation.

Is going from "deworming children make less poor" to "the worms cause poor outcomes" such a stretch?

Sure, it is likely.

[doukdouk]

All those many factors equally affect both groups, that's the point of running a randomized trial. I am not sure how they are relevant to the claims that worms cause poor outcomes or that deworming children make them less poor adults. Could you please explain?

[peter_retief]

My problem is how do you run a randomized trial with something as common as a deworming pill that is handed out in clinics by nurses and can be bought over the counter? I am just curious.

[doukdouk]

I don't see how that would be a problem: even if it was common, handing out pill to one group but not the other would still lead to one group being more "dewormed" than the other, all others things being equal.

Anyway, in this case, I do not believe deworming was as common as you assume:

> Baseline parasitological surveys indicated that helminth infection rates were over 90%, and over a third had a moderate-heavy infection according to a modified WHO infection criteria (Miguel and Kremer 2004)

> Drug take-up rates were high, at approximately 75% in the treatment group, and under 5% in the control group (Miguel and Kremer 2004).

5% of the population taking the drug when >90% should means the drug is not "common".

[peter_retief]

In South Africa it has always been the default to deworm all children. In schools, clinics. https://www.sanews.gov.za/south-africa/government-launches-d...

[doukdouk]

This trial was done in Kenya, not South Africa.