Jeff Perry originally published his breakthrough article strongly suggesting the ability of a CT to rule-out subarachnoid hemorrhage (SAH) within six hours of onset in 2011. Since then a few papers added some data supporting the conclusion's of the Ottawa group, including a 2015 paper from Netherlands obtaining very similar results in a rather community-based practice.
Recently; Nicole Dubosh and Mayo's Fernanda Bellolio along with colleagues published a systematic review synthesizing the evidence of five papers and around 8000 patients. The paper has a lot of very interesting technical and clinical pearls that are important to read, but in summary if a patient presents with a history suggestive of SAH and a CT is performed within six hours of onset by a neuroradiologist, the sensitivity of the imaging is 98.7%, specificity is 99.9% and the pooled likelihood ratio of a negative CT was 0.010. In simpler terms, if conditions met, the chances of missing a SAH are about 1/1000.
Here the link to the full paper.