Effect of home zinc on rate of sick visits to Lwak clinic, hospitalization, and mortality, western Kenya, from February 2008–March 2009, controlling for baseline rates of morbidity in home zinc and comparison villages.

Intention-to-treat analysis includes all enrolled children. Rates are given as episodes per person-year.*Pre-intervention period was October 2006 to November 2007.†While pre-intervention includes all children in the village of appropriate age, the intervention period only includes those enrolled in the home zinc study.‡Rate ratio is comparing rates between home zinc and comparison groups during the intervention period, adjusted for pre-intervention rates of same syndrome in the child’s village, distance of child’s compound to Lwak Hospital and child’s age.¶For primary outcome, intracluster correlation (ICC) = 0.027.