Sleep doesn't evade older adults because of aging alone, researchers affirmed in a Chinese study with the largest-ever cohort of the very elderly. Centenarians in China were 70 percent more likely to report good quality sleep than younger seniors age 65 to 79, according to Danan Gu, of Portland State University in Portland, Ore., and colleagues. All other factors being equal, good sleep quality was 19 percent more common among respondents in their 80s and 38 percent more likely among those in their 90s, compared to those from age 65 to 79. "These findings may support the argument that sleep problems at old and oldest-old ages likely arise from a variety of physiological and psychosocial factors rather than aging per se," Gu's group wrote in the May 1 issue of SLEEP. Another possibility is that people change their perception of "acceptable" sleep with advancing age, they suggested. The researchers analyzed a nationally representative, population-based survey dataset from mainland China, which has the world's largest population of nonagenarians and centenarians. The 2005 wave of the Chinese Longitudinal Healthy Longevity Survey included interviews with 15,638 individuals age 65 and older (43 percent men, 57 percent women). Their ages broke down as follows: * 5,047 age 65 to 79 * 3,870 in their 80s * 3,927 in their 90s * 2,794 age 100 or older Overall, 65 percent reported good sleep quality, with 7.5 to 8 hours being the most common duration. Sleep quality didn't appear to decline with age, but there were trends for better sleep. Men were 42 percent more likely to report good sleep than women. Those of Han ancestry were 34 percent more likely to report good sleep than ethnic Chinese minorities, and those in rural areas were 9 percent more likely to report better sleep than urban dwellers. After adjustment for socioeconomic condition, family or social support, smoking and drinking habits centenarians were 13 percent more likely to report good sleep quality than their 65- to 79-year-old counterparts. Higher family economic status and adequate access to health care were more important factors, increasing the odds of good sleep quality by 67 percent and 120 percent, respectively. Education, marriage, and number of living children didn't impact sleep quality. But living with a spouse or family member increased the odds of good sleep quality by 11 percent compared with living alone. Institutionalization decreased the odds of good sleep by 22 percent. Paradoxically, quality sleep was 13 percent more likely among current smokers than nonsmokers and 27 percent more likely among those who drank alcohol than among nondrinkers. Researchers said this might be explained by the fact that these two groups were relatively healthy within the cohort. Adding in the impact of health did not change these patterns. "In other words, poor health did not account for the associations between sleep duration and most other variables, except for age," Gu's group wrote in the paper. For sleep duration, though, poor health tended to increase the odds of reporting too little sleep (six hours or less) or too much (10 hours or more). The researchers cautioned that the study could not draw causal relationships, and that it included only self-reported data without a measure of use of sleep medication.