In Reply: Drs Gianakos and Mehra raise concerns that participants in our study may have already developed mild cognitive impairment at baseline. The cognitive test scores presented in Table 1 in the article were results from the shortened version of the MMSE, which is scored out of 26 points, rather than the 30-point scale of the full MMSE. Although we cannot be sure that participants in our cohort did not have some subtle preclinical cognitive impairment at baseline and that this was greater among women with sleep-disordered breathing, the 2 groups had very similar mean (SD) test scores: 24.9 (1.2) for women without sleep-disordered breathing and 25.1 (1.1) for those with sleep-disordered breathing (P = .22). In the analysis, we also addressed this issue by excluding any suspected cases of cognitive impairment before the 5-year follow-up cognitive assessment, defined as those participants with a physician's diagnosis of dementia or Alzheimer disease or a low cognitive test score. In addition, when dementia and mild cognitive impairment were analyzed separately, the associations with sleep-disordered breathing were similar, albeit with reduced power (unadjusted odds ratio for mild cognitive impairment or dementia, 1.80 [95% CI, 1.10-2.93]; for mild cognitive impairment, 1.88 [95% CI, 1.04-3.40]; for dementia, 1.70 [95% CI, 0.88-3.27]).