Comparison of text-messaging to voice telephone interviews for active surveillance of adverse events following immunisation.