In Reply: Dr Tsai's comments reflect a desire to hold epidemiological research in settings of political violence and human rights abuses to the highest standard. We agree with this goal because only through rigorous research can evidence-based interventions be developed with the most effective and appropriately targeted use of resources.
We share Tsai's concern regarding possible unobserved covariates associated with preexisting differences between child soldiers and never-conscripted children. Although we used matching, it is a partial solution. An alternative would be the use of propensity score matching with large sample sizes. However, we caution against researching violence and human rights solely with epidemiological approaches dependent on large samples, which could also misdirect funding in resource-poor settings and potentially overestimate disorder rates without well-funded validation studies.1
Kohrt BA, Jordans MJD, Worthman CM. Measuring Mental Health in Child Soldiers—Reply. JAMA. 2008;300(23):2729-2730. doi:10.1001/jama.2008.775