I think a wide CI means your result is not that accurate/reliable. For example because you used a small sample size. The more narrow your CI, the more your result from your sample will correspond to the true result (complete population).
I think if it contain 1; than it's not significant