Background The oldest-old (≥80 years) are the most rapidly growing age group globally, and they are most in need of health care and assistance. The aim of this study was to explore whether increased longevity is leading to populations of healthier oldest-old (compression of disability; benefits of success), or to less healthy and disabled oldest-old (expansion of disability, costs of success). Methods Data are from Chinese Longitudinal Healthy Longevity Study (CLHLS). Three pairs of cohorts aged 80–89, 90–99 and 100–105 (in total 19,528 oldest-old participants) were examined; the two cohorts in each pair were born ten years apart, with the same age at the time of the assessment in the CLHLS 1998 and 2008 surveys. Findings Controlling for various confounding factors, compared to cohorts born 10-years earlier, death rates at oldest-old ages among later cohorts were substantially reduced annually by 0.2%–1.3%, and their disabilities of Activities of Daily Living had significantly reduced annually by 0.8%–2.6%; however, the later cohorts’ cognitive impairment rates increased annually by 0.7%–2.2% and objective physical performance capacity (stand-up from a chair, pick-up a book from floor, turning around 360°) decreased annually by 0.4%–3.8%. We also found that mortality of female oldest-old was substantially lower, but their functional capacities in ADL, cognition and physical performance were worse, compared to their male counterparts (mostly P<0.001). Interpretation These empirical findings can be explained by the theoretical framework of mixed effects of two opposing processes: advances in medications, lifestyle and socioeconomics may compress ADL disability, that is, “benefits of success”, but lifespan extension may expand disability of physical and cognitive functioning as more frail elderly survive with health problems, that is “costs of success”. Recent improvements in living standards and availability of facilities for daily living among Chinese elderly may also contribute to the contrasting trends of ADL disability and physical performance.

In October, 2015, China’s one-child policy was replaced by a universal two-child policy. The effects of the new policy are inevitably speculative, but predictions can be made based on recent trends. The population increase will be relatively small, peaking at 1·45 billion in 2029 (compared with a peak of 1·4 billion in 2023 if the one-child policy continued). The new policy will allow almost all Chinese people to have their preferred number of children. The benefits of the new policy include: a large reduction in abortions of unapproved pregnancies, virtual elimination of the problem of unregistered children, and a more normal sex ratio. All of these effects should improve health outcomes. Effects of the new policy on the shrinking workforce and rapid population ageing will not be evident for two decades. In the meantime, more sound policy actions are needed to meet the social, health, and care needs of the elderly population.

In the absence of manipulation, both the sex ratio at birth and the population sex ratio are remarkably constant in human populations. Small alterations do occur naturally; for example, a small excess of male births has been reported to occur during and after war. The tradition of son preference, however, has distorted these natural sex ratios in large parts of Asia and North Africa. This son preference is manifest in sex-selective abortion and in discrimination in care practices for girls, both of which lead to higher female mortality. Differential gender mortality has been a documented problem for decades and led to reports in the early 1990s of 100 million ''missing women'' across the developing world. Since that time, improved health care and conditions for women have resulted in reductions in female mortality, but these advances have now been offset by a huge increase in the use of sex-selective abortion, which became available in the mid-1980s. Largely as a result of this practice, there are now an estimated 80 million missing females in India and China alone. The large cohorts of ''surplus'' males now reaching adulthood are predominantly of low socioeconomic class, and concerns have been expressed that their lack of marriageability, and consequent marginalization in society, may lead to antisocial behavior and violence, threatening societal stability and security. Measures to reduce sex selection must include strict enforcement of existing legislation, the ensuring of equal rights for women, and public awareness campaigns about the dangers of gender imbalance.Asia ͉ sex ratio ͉ son preference