A total of 12 trials have tested the effect of neonatal vitamin A supplementation (NVAS) on mortality. Overall, NVAS had no effect on mortality, but results were heterogeneous. Two competing hypotheses have been put forward to explain the divergent effects: A) NVAS works by preventing vitamin A deficiency (VAD) and not all countries have VAD; B) NVAS interacts negatively with subsequent diphtheria-tetanus-pertussis (DTP) vaccine, increasing mortality in females; in countries with low DTP coverage NVAS may have a beneficial effect. Only hypothesis A was tested in a recent meta-analysis; there is no strong empirical support for hypothesis A and it would not explain observed negative effects in some settings. Hypothesis B accounts for most observations. However, so far it has only been tested properly in a few trials. If hypothesis B is correct, it has major consequences for the understanding of the effects of vitamin A, and for the VAS policy in older children. As a WHO priority, the DTP coverage is bound to increase, and therefore hypothesis B urgently needs to be tested.