A proactive response to population aging is a key factor in determining a country's long-term economic growth momentum [23]. The health-related losses associated with an aging population not only place a heavy physical, mental, and economic burden on individuals, but also put a constant strain on healthcare systems. Promoting the development of a healthy aging system is, therefore, of great value [24]. Identifying the major health losses due to population aging is important for developing successful response strategies. Given this challenge, this study combined two health indicators, HALE and DALY, and proposed an attribution analysis framework to systematically investigate how population aging affects the quality of life and disease burden. We applied this framework to regional populations to comprehensively analyze the health impacts of population aging from two perspectives. To the best of our knowledge, this is the first study to quantitatively attribute and predict shifts in epidemic patterns of disease and injury due to population aging. Our proposed framework serves as an important top-level design to guide practice, allowing comparisons of the health effects of population aging across different time periods and regions.

This study found that the decline in premature deaths in both sexes was the main determinant of HALE growth during the period 2010 to 2020, with people aged 60 years and over contributing more than half of this growth. This finding aligns with the national-level trends observed from 1990 to 2013, where improvements in HALE were similarly attributed to reductions in premature mortality among older adults [19]. With the transition in population health, the leading cause of the improvement in death levels has shifted from a significant increase in newborn survival to a significant increase in life expectancy among older adults, who now play a major role in extending both life expectancy and HALE [25]. However, this study also highlights that the positive effect of improved disability levels on the growth of HALE remains minimal, resulting in HALE failing to keep pace with the growth rate of life expectancy. This growing disparity indicates that longer life expectancy increasingly coincides with illness or disability, intensifying the demand for healthcare and long-term care among the elderly. This trend imposes substantial  economic costs and intangible burdens on individuals, families, and society [26]. In the long term, mortality reduction tends to decelerate as it approaches a lower threshold. For regions with lower baseline population health, higher mortality theoretically has more potential for improvement and may achieve a faster rate of decline. In more socially and economically developed regions, mortality may reach a  plateau that is challenging to lower further [27]. Comparing the results of this study with the gross mortality from all diseases and injuries in China reported by the GBD, the mortality in Guangzhou was significantly lower than the national level during the study period. However, the rate of decline was also slower than the national level, suggesting that the future causes of population health in advanced regions may gradually shift from death to disability.

The projection of future HALE indicates a steady increase in HALE for both sexes over the period 2020 to 2030. During this period, the reduction in premature deaths among people aged 85 years and over is expected to become the main contributor to the all-age mortality effect. However, the relative reduction in premature deaths among other age groups will limit the all-age mortality effect, preventing it from exceeding historical trends. This phenomenon may align with the relationship between life expectancy and mortality, which follows a logistic curve. As  life expectancy improves, the rate of increase gradually slows and eventually approaches zero, stablizing at a fixed constant. Based on this principle, some researchers have proposed that there is a natural physiological limit for human beings in the absence of external risk factors. However, the onset of chronic diseases can be delayed or even prevented by adopting health lifestyle choices such as reducing tobacco and alcohol consumption, maitaining a healthy weight, increasing physical activity and so on. These measures can compress the period between the onset of disease and death [28]. The findings of this study are also consistent with these theories. If the mortality effect for both sexes does not increase significantly, the proportion of the disability effect of all ages and the elderly population in the overall effect will increase. Notably, the contribution of disability improvement among elderly women to the growth of HALE shifts from negative to positive, suggesting that a future trend of morbidity compression in advanced regions, where the quality of life for older populations is likely to improve.

This study also found that the increase in DALY attributable to population size growth and age structure changes for both sexes between 2010 and 2020 outweighed the decrease from all other factors, resulting in a net increase in total DALY. Population size growth was the main driver of the increase in DALY, similar to the global trend in previous studies [29, 30]. Ongoing population change is one of the hallmarks of the fourth stage of epidemiology, in which chronic degenerative diseases are the main health threats. Improvement in disease prevention and control technology, as well as other external population factors, are delaying the onset of disease, reducing post-disease mortality and increasing the proportion of older people in the population age structure. The size of the population only increases over time, so the aging of the population is the main contributor to the accumulation of the disease burden [31]. Data from the National Population Census indicate that China began entering the fourth stage of health transformation during the period 2010 to 2020, which has been influenced by macro- and micro-level factors such as the continuous promotion of healthy aging policies, the gradual improvement in family and community environments suitable for aging populations, and the continuous improvement of people's health awareness and lifestyles [32]. This study also confirmed that under the guidance of the Chinese central government's top-level strategies to actively manage population aging, the impacts of population aging on the burden of disease in developed regions reflected the characteristics of the fourth stage of health transformation. Besides, DALY attributable to age structure accounted for a substantial proportion in both sexes, second only to the contribution of population growth.

The projection of future DALY suggests that the net increase in DALY for both sexes in the period 2020–2030 will exceed the historical trends, with total DALY continuing to accumulate over time. Unlike HALE, the accumulation of DALY is influenced by both population changes and the prevalence intensity of different causes, meaning there is no upper limit to DALY accumulation. Consequently, the healthcare system will face an increasing burden of disease. In addition, the DALY attributable to age structure changes will surpass not only the effects of population growth but also all other causes, suggesting that age structure change has transitioned from being a secondary to a primary driver of DALY accumulation.

The attribution of population aging to HALE and DALY in this study can be interpreted through two complementary aspects of health measurement. Attributed HALE examines the per capita health loss in a hypothetical population. It is a health expectancy measure calculated by applying the real-world prevalence intensity of causes, expressed as a percentage of the disease spectrum, to a hypothetical cohort within the life table. The attributed HALE, therefore, highlights that improving the health of the older population can effectively extend the average healthy lifespan. The attributed DALY examines the cumulative loss of health in the real population, which accrues as the age structure shifts toward an older demographic. Attributive DALY, thus, represents the excess burden of disease that must be managed and controlled by the healthcare system during the period. The attributed HALE identifies the prevalence intensity of specific causes and their health effects from the perspective of the elderly population, while the attributed DALY identifies the impacts of changes in age structure of the population on the burden of disease from the perspective of societal development.

Empirical evidence indicates that the five leading causes of disease burden attributable to changes in age structure are all non-communicable diseases (NCDs). This trend is projected to persist in the future. The primary prevention of NCDs requires the improvement in population health literacy and the creation of a health-promoting environment [33, 34]. The efficacy of primary prevention is contingent upon the willingness of the population to engage in preventive measures [35]. For the elderly, offline intervention and health support tools are more suitable [36]. A health-promoting lifestyle can also improve the condition of elderly individuals with NCDs and delay the onset of complications [37]. The establishment of health records and the provision of medical follow-up services by community health centers represent effective health-promoting lifestyle interventions [38]. In addition, the burden of several interrelated diseases is projected to increase substantially, and the co-occurrence of these diseases (multimorbidity), such as nutritional deficiencies and musculoskeletal disorders, must be taken into account [39]. Aging and degeneration of the digestive system can lead to reduced nutrient intake or utilization, which in turn increases the risk of musculoskeletal disorders [40, 41]. Integrating targeted dietary advice and nutritional support into the musculoskeletal treatment program is an effective way to reduce the comorbidity of these two diseases [42]. Multimorbidity usually involves different combinations of risk factors or pathobiological mechanisms, which may alter the benefits and effectiveness of health interventions [43, 44]. An evidence-based, fundamental restructuring of interventions for older people with multimorbidity is essential to ensure their full integration into active aging practices [45].

A major limitation of this study is that it did not quantify the intensity and direction of the effects of population aging on the burden of multimorbidity. The probability of related diseases co-occurring is often much higher than the product of the probabilities of individual diseases, and the burden of multimorbidity may also be higher than the sum of the burdens of the individual diseases. However, this study included complete categories of causes and the findings hightlight key contributors to a substantial burden of disease, which may also overlap and interact. Given that the essence of the burden of disease is to obtain comparable relative importance of different causes using a consistent measurement framework, this study can still provide conservative estimates of the burden of different diseases and injuries, indicating the potential patterns of multimorbidity, and providing an important decision-making basis for formulating public policies to actively cope with population aging.