By Dr W.A. Arrindell, psychologist and consultant

Background

Studies have demonstrated that good sleep may increase one's life expectancy. Most of these studies, however, examined this association in subjects with a limited age range (usually 30-65 years). Such studies do not provide an answer to the question as to whether this association also holds for the oldest-old, ages 80 years and older. A recent study conducted in China by Dr. Danan Gu and colleagues is the first to examine sleep issues in a very large sample of exceptionally old adults, including as many as nearly 2,800 individuals who were 100 years of age and older (SLEEP 2010; 33[5]: 601-610).

The survey

The study used data from the 2005 wave of the Chinese Longitudinal Healthy Longevity Survey. It began in 1998 and is the first nationwide longitudinal survey focusing on the oldest-old ever conducted in a developing country. In-home interviews were conducted in 22 provinces in mainland China. A total of 15,638 individuals ages 65 and older were interviewed, including 5,047 young seniors ages 65 to 79, and 10,591 oldest-old seniors ages 80 and older. The subsample of the oldest-old comprised 3,870 subjects ages 80-89 years (=octogenarians), 3,927 persons between 90 and 100 years old (=nonagenarians), and 2,794 people who had attained the age of 100 years or higher (=centenarians); 6,688 respondents were men, and 8,950 were women.

In addition to data on sleep quality ("How do you rate your sleep quality recently?") and hours of sleep ("How many hours on average do you sleep every day including napping?"), Dr. Gu and co-workers also collected data on background demographic characteristics (e.g., marital status, income, urban versus rural residence), lifestyle (including smoking, alcohol use), health and psychological characteristics, to mention but a few.

Findings

The majority (sixty-five percent) of Chinese elders reported good sleep quality. The average number of self-reported hours of sleep was 7.5 hours.

Males, people living in rural areas, people who had optimal access to healthcare, those who had a high socioeconomic status (the resource-rich), and subjects in good health reported better quality of sleep than, respectively, females, people living in urban areas, people who had suboptimal access to healthcare, those who had a low socioeconomic status (the resource-poor), and subjects in poor health. All other factors being equal, thus holding constant factors that might blur the association between sleep quality and age, octogenarians, nonagenarians, and centenarians were more likely to report good sleep quality than the young elders ages 65-79 years.

Pointing to the importance of the availability of family/social support, marital status was positively associated with sleep duration, in that married elders were more likely to report an average duration of sleep (7-8 hours) than their unmarried or no longer married counterparts.

Oldest-old adults ages 80 years or older tended to sleep either shorter (6 hours or less) or longer (10 hours or more) than young elders. According to the researchers, this is because when old adults grow older their health condition tends to get worse, on average. Indeed, their data demonstrated that poor health conditions tended to increase the likelihood of reporting either too little (6 hours or less) or too much (10 hours or more) sleep. Importantly, when health conditions were taken into account (held constant), nonagenarians and centenarians were less likely to sleep too little than young elders ages 65-79 years.

Conclusions

Age and health conditions are the two most important factors associated with self-reported sleep quality and sleep duration. Good quality of sleep among long-lived old adults may have some implications for achieving healthy longevity.

Interpretation

The nature of the study (associations were determined at one and the same point in time in a so-called cross-sectional approach) does not allow drawing the conclusion that good quality sleep leads to healthy longevity: the better one's sleep quality and duration, the longer one will live in good health (a causal relationship). The findings do demonstrate that good sleep quality and duration coincide (co-occur) with healthy longevity, which implies any of three interpretations: (a) healthy, very old people sleep well and sufficiently because of their health condition; (b) optimal quality and duration of sleep precede and predict future healthy longevity, or (c) both (a) and (b) are true, that is the relationship is bi-directional.

For a causal interpretation of the findings, a prospective-longitudinal (follow-up) approach is needed where sleep patterns studied at younger ages (time-point 1) are used to determine whether they predict healthy longevity years later (time-point 2), even after holding constant factors that also influence the relevant relationship. This is acknowledged by Dr. Gu and associates with noting that "once the 2008 [follow-up] wave data are released, we will extend our current preliminary results to examine how sleep quality and duration in 2005 are associated with health conditions in 2008 and vice versa" (p. 608).