Health is the key to the aging problem, and solving health problems can essentially resolve the negative impact of aging[1, 2]. Havighurst first proposed the concept of "healthy aging". In 1990, the first World Assembly on Aging officially introduced the strategy of "healthy aging" to its member nations. The strategy called for enhancing or maintaining the internal abilities of elderly individuals and improving supportive environments to realize the functions required for the healthy life of elderly people. China has the largest number of elderly people in the world, some scholars have pointed out that healthy aging is the key way for China to cope with the challenges of population aging[5–7].
"Healthy aging" is the overall changing trends in the health of all elderly individuals in a society[1, 8]. China was recognized as an aging society in 1999, and by 2019, elderly individuals over 60 years old accounted for 18.1% of the total population and those over 65 years old accounted for 12.6%. In the past 20 years, has China moved towards healthy aging, sub-healthy aging or even diseased aging? According to the Healthy China Action Promotion Committee, in 2018, China's average life expectancy was 77 years, of which the healthy life expectancy was 68.7 years; that is, elderly individuals lived with diseases for approximately 8.3 years, including 40 million half-disabled elderly individuals and 20 million completely disabled elderly individuals, and 180 million elderly people with one or more chronic diseases, accounting for 75% of this population. In addition to the above officially disclosed data, researchers have also carried out a rich exploration and assessed the changing trends in the health status of the elderly population in China. Some of them have found that the elderly population is getting healthier[9–11], while some have reached opposite conclusions[12–14]. In addition, some of the studies have been positive for some indicators and negative for other indicators, without consistent results.
These annual statistical indicators and research literature studies are primarily based on cross-sectional data in a certain year. These measurement results can only reflect the static level of elderly health at that time point while cannot describe the long-term changes. Individual longitudinal research studies based on multistage data often compare only the measurement results in each stage but do not test the significance of the differences between measurement values across the periods. Therefore, whether the changes across time periods show statistical significance has not been tested. Most of the measurement tools use one or several indicators, which are often separately measured in operation. A single or a small number of indicators are not enough to reflect the whole picture of the health of the elderly population, leading to one-sidedness and uncertainty. However, due to different research methods and variables used in the construction of indicator systems, most of the results lack comparability[17, 18]. These shortcomings obviously affect the accurate understanding of the overall characteristics and trends in aging health in China and may be the main cause of the inconsistencies in the existing research results.
In this paper, improvements over previous studies will be made: using 5 phases of longitudinal data, the "overall health" of elderly individuals will be measured with the four dimensions of physical health, functional status, mental health and social health, and statistical test methods will be used to test the significance of the difference across the 5-phase measurement results of the average health in the elderly population. Based on the results of these dynamic measurements and tests, the trends in healthy aging in China will be judged.
In response to the challenges of population aging, Havighurst first proposed the concept of "healthy aging", which was defined as prolonging life span and increasing life satisfaction. However, longevity reflects only the quantity of life, not the quality of life. In 1987, the World Health Assembly extended the definition of "healthy aging" as postponing biological aging and social aging through a series of positive measures at the same time as the unstoppable calendar aging continues. Rowe and Kahn defined healthy aging as an "active disease-free state". These definitions ignore the inevitable aging process in the process of life, and it is not realistic to expect elderly individuals to have no disease. In view of this, in 2016, the World Health Organization no longer emphasized the lack of "disease" in its global report on aging and health but defined "healthy aging" as "the process of developing and maintaining the functions required for healthy life of the elderly" based on the perspective of "function". This definition mainly depends on older people's internal abilities, supportive environment and their interaction.
The concept of "healthy aging" based on functional performance is widely accepted, but its measurement indicators, statistical analysis methods and models are still very limited[20, 21]. This is because measuring healthy aging is not simply equivalent to measuring the health of elderly individuals. There are two large differences in measuring healthy aging: first, the health of elderly individuals is the result of the gradual development of people across the whole life process. Individuals’ choices at different time points and environmental interventions during the life process will change their internal abilities and function and ultimately affect the changing trajectory of healthy aging. Therefore, we should dynamically investigate or intervene in healthy aging based on the whole life course; second, although healthy aging should be based on individual's health and longevity, it focuses on the overall situation or average level in the elderly population. The healthy life span of a few people has little effect on improving the average healthy life span of the group. Xiong & Dong, Wu & Jiang and Zhong & Chen pointed out that "healthy aging" can be considered from two aspects: first, healthy life expectancy is increasing, the corresponding period of disease is shortening, and the best goal is to "end without disease"; second, the proportion of healthy and long-lived elderly people accounts for the majority and is increasing. It can be seen that healthy aging is based on the measurement of the overall health level of the elderly population and then the trends in the dynamic changes.
The most commonly used indicators of elderly health include life expectancy, health expectancy, quality-adjusted life expectancy (QALE), disability-adjusted life expectancy (DALE, which was renamed HALE in 2001), years of life lost (YYL), incidence rate, disability rate and other indicators. In research studies, some researchers will use a single indicator, while others will use multiple indicators.
Researchers often measure the health of elderly individuals from different dimensions according to the definition of health from the World Health Organization. The measurement of physical health mainly includes the activities of daily life (ADL) scale, Barthel Index (BI) scale, instrumental ADL (IADL) scales and so on. Subjective well-being, a comprehensive dimension, is commonly used in the assessment of mental health of elderly individuals and includes life satisfaction, positive emotions, negative emotions, etc.. These scales mainly include happiness scales (e.g., Memorial University of Newfoundland Scale of Happiness; MUNSH), loneliness scales (e.g., the UCLA loneliness scale), depression scales (e.g., Center for Epidemiological Studies Depression scale; CESD), the Life Satisfaction Index A scale, the Mini-Mental State Examination (MMSE), life significance scales (e.g., Meaning of Life Questionnaire; MLQ), etc. Assessments of social health often use social adaptation, social participation, social role, social support and other indicators. The scales include the earliest Berle Index of social support, social relationship scale (SRS), social support questionnaire (SSQ), social adaptation scale (SAS), social adaptation self-test scale (SAS-SR), and the social maladjustment questionnaire (SMS).
It is difficult to reflect the overall health status of elderly individuals with only a single indicator or a single dimension. Measurement errors often lead to the illusion that elderly individuals suffer from group diseases or functional problems[18, 32]. Therefore, currently, researchers have measured the health of elderly individuals from three dimensions of physical health, mental health and social health. When elderly individuals are asked about themselves, they think that healthy aging should include four aspects: physical, mental, psychological and social. In recent years, more emphasis has been placed on the comprehensive evaluation of the health of elderly individuals based on the concept of overall health including physiological health, physical function, mental health, role function, self-assessment health and other dimensions, as well as health-related quality of life assessment (HRQOL), comprehensive geriatric assessment (CGA), etc[34, 35]. A developed comprehensive assessment scale includes the Duke University Older Americans Resources and Services (OARS) assessment scale, the Comprehensive Assessment and Referral Evaluation (CARE), the Frailty Index (FI), and the Grade Membership of Health Status. Compared with traditional measurement methods, the comprehensive assessment of elderly health as a predictor of elderly health is much better, which can provide a more appropriate entry point for public health policy[4, 18].
It can be seen that the measurement of elderly health has experienced a development process from a single indicator to multiple indicators and from a single dimension to multidimensional to a comprehensive evaluation. The same process also stands for the research on the measurement of elderly health in China. Zhong & Chen and Qiang & Zhe used the Sullivan method to measure the healthy life expectanc y of Chinese elderly individuals. Wu et al.proposed measuring the health of Chinese elderly individuals with three indicators: life span, self-care ability evaluations and self-assessed health. Raoselected four indicators to measure the social effect of healthy aging, including the Chinese elderly individuals' life independence, spiritual pleasure, social interaction and social contribution. In fact, these studies all use single indicators for the measurement. Jiao K  made a contribution in measuring the overall health of elderly individuals by using the ADL, IADL, MMSE and self-reported measures of chronic diseases and used a latent class model to classify the Chinese elderly participants into four categories: healthy, mild disability, cognitive impairment and mostly unhealthy. Based on the four dimensions of life independence, spiritual pleasure, social interaction and participation, and social contribution, Qian J found that China's healthy aging has achieved some results, but there is still much room for improvement, especially focusing on the health status of the elderly population in rural areas. These researchers have tended to understand healthy aging as a measurement of aging health and attempt to reflect trends in healthy aging from the results of cross-sectional data from a particular year. Obviously, the static measurement results have difficulty reflecting long-term trends in healthy aging in China.
Some researchers have realized that we should judge trends in healthy aging in China from the perspective of dynamic change. Mu G  proposed a set of eight indicators including the change in proportion of healthy elderly in the total population and the change in proportion of healthy elderly in the total elderly population to reflect health changes in the elderly population in China, but the data were not used to conduct specific measurement work. Zhe T, Xiang M and Fang X used ADL scores to track 12-year survey data from Beijing elderly individuals, which showed that the ratio of healthy life expectancy (ALE/LE) decreased in recent years. A study from the World Health Organization found that the gap between the average life expectancy and the healthy life expectancy of elderly individuals in China increased with increasing age in the period from 2000 to 2012. Zeng Y and Feng Q showed that cognitive function and physical function of Chinese elderly individuals significantly decreased from 1998 to 2008 compared with 10 years ago. The increase in life expectancy per capita was accompanied by a decline in the health level of elderly individual. In contrast to the conclusions of these studies, other studies found that Chinese elderly individuals are becoming healthier and healthier[9–11]. Yu Y and Feng J , based on the data of the China Nutrition and Health Survey (CHNS) in 1991–2009, using a random effect model, found that indicators of daily behavioral abilities gradually improved between generations, but the indicators of chronic disease and health risk gradually deteriorated and was more severe in rural areas. These studies have focused on dynamic measurements of healthy aging, but their shortcomings are that they also examined a single health indicator, or a small number of health indicators, and the empirical results were not consistent.
From the perspective of measurement research on healthy aging in China, the existing problems are that the measurement indicators are too few, the dimensions are single, and the results are not consistent. In addition, most of these studies were aimed at static measurements of cross-sectional data at a particular period of time or lack statistical tests assessing the significance of difference between measures across multiple periods of data. These shortcomings have limited judgments regarding trends in healthy aging in China.