Average tuberculous infection prevalence estimates by intervention arm based on estimated prevalence calculated by various published methods.

+These estimates are based on five and not six prevalence estimates as all other estimates in the table. This is because for two communities we could not clearly define a “true” mode of TST induration. Both these communities have low prevalence estimates as calculated by all other methods and would most probably produce low infection prevalence estimates for the mirror method too. We believe the observed differences in between intervention arms averages are an artefact of what we describe here.