Comparisons between the self-assessed and observer-assessed presence and severity of colds