Some problems with evidence-based aid

In Against Health: How Health Be­came the New Mo­ral­ity (Biopoli­tics, Medicine, Techno­science, and Health in the 21st Cen­tury) , there’s an es­say “Against Global Health?” by Vin­canne Adams.

The in­tro­duc­tion was a bit of a slog, but I think the point was that if “health” is defined as some­thing which can and should be given to peo­ple re­gard­less of what they think in the mat­ter, the re­sults can be pre­sump­tu­ous.

While she may be overly in­vested in the way things are usu­ally done, she brings up some dis­quiet­ing points about the ways in­sist­ing on dou­ble-blind tests can go wrong. For ex­am­ple, she men­tions a pro­ject to eval­u­ate train­ing in safe in­fant de­liv­ery tech­niques in Ti­bet which was scut­tled be­cause not enough women were dy­ing to get a good power calcu­la­tion.

She de­scribes ex­per­i­ments which are done in iso­lated com­mu­ni­ties be­cause that’s where the re­searchers can be rea­son­ably sure that the sub­jects don’t have al­ter­na­tive sources of care which would foul up the dou­ble-blind­ing. This does seem like a drunk and lamp post prob­lem. [1] More gen­er­ally, the con­cern is that ac­tual care is de­layed in the search for perfect in­for­ma­tion. Ad­mit­tedly, it’s hard to be sure about how to bal­ance search­ing for in­for­ma­tion with tak­ing ac­tion, but it strikes me as a prob­lem that’s worth some thought rather than just as­sum­ing that dou­ble-blind­ing is a re­li­able im­prove­ment.

[1] Is there a stan­dard LW term for search­ing where it’s easy to search rather than where the an­swer is likely to be?