Skip to main content

Trajectories of middle-aged and elderly people’s chronic diseases Disability Adjusted Life Years (DALYs): cohort, socio-economic status and gender disparities



The accelerated aging trend brought great chronic diseases burdens. Disabled Adjusted Life Years (DALYs) is a novel way to measure the chronic diseases burden. This study aimed to explore the cohort, socioeconomic status (SES), and gender disparities of the DALYs trajectories.


A total of 15,062 participants (55,740 observations) comes from China Health and Retirement Longitudinal Study (CHARLS) from 2011 to 2018. Mixed growth curve model was adopted to predict the DALYS trajectories in 45–90 years old people influenced by different birth cohorts and SES.


We find significant cohort, SES (resident place, education level and income) disparities differences in the chronic diseases DALYs. For individuals of earlier cohort, DALYs are developed in a late age but grow fast with age but reversed for most recent cohorts. Living in urban, having higher SES level will decrease the growth rate with age, but converges for most recent cohorts. Meanwhile, DALYs disparities of resident place and education level show gender differentials that those for female are narrowed across cohort but for male are not.


The cohort effects on chronic diseases DALYs are accumulated with China’s unique social, and political settings. There are large inequalities in early experiences, SES and DALYs. Efforts of reducing these inequalities must focus on the lower SES individuals and those living in rural areas, which greatly benefit individuals from recent cohorts.


Chronic disease problem of middle-aged and elderly people is an increasing serious public health challenge in China that places a heavy burden on health care [1]. According to China Center for Disease Control and Prevention (CCDC), 75.8% of the elderly population of 60 years and older in China are troubled by one or more chronic diseases, which cause a lot of physical and psychological harm to the patients due to the long course of the disease and the protracted condition, and decrease the life years [2].

Understanding the change trend of chronic diseases with age is a critical way to control them. In the former researches, morbidity and mortality are the mostly used indicators to measure the diseases [3,4,5,6]. But after Global Burden of Disease Study 2010 (GBD study 2010) published the measurement of Disabled Adjusted Life Years (DALYs) of 291 diseases and injuries, DALYs become a novel indicator to evaluate the chronic diseases. Since then, there will be measurements of chronic diseases by DALYs being published every year. However, most of them focused only one disease and analyses of the disease lack of longitudinal perspective.

As a novel health indicator, DALYs of chronic diseases may also show cohort disparities that was demonstrated having significant effects on the growth trajectories of health indicators like Self-rated Health (SRE), mental health index, Body Max Index (BMI) and index of Activities of Daily Living (ADL) [7,8,9,10,11]. Birth cohort is usually the represent of life course in former studies. Last century, China has witnessed a series of dramatic political, economic, and cultural upheavals, including the war (1937–1948), the great famine (1958–1961) and Cultural Revolution (1966–1976). Individuals with different birth cohort usually has different life courses, this suggests the chronic diseases DALYs of these middle-aged and elderly people may also show cohort disparities.

Besides cohort effects, previous studies also found individuals with higher socioeconomic status (SES) for example owning higher education level, living in urban area and having higher income are more likely to report better health [12,13,14], greater levels of physical functioning or mobility [15,16,17], better mental health outcomes [18,19,20], and lower rates of disability and mortality [21,22,23]. For chronic diseases, since they are closely associated with lifestyle factors like smoking, drinking and poor dietary habit [24], higher SES may have an adverse effect. When put in the context of age and life course, associations of SES and health were found gradually changed with age and interacted by cohort effect. The impact of different levels of education and income accumulates over the course of a lifetime, resulting in inequality in the health of the elderly. Zhu et al. found the advantages brought by higher education level were lager with age [25]. Chen et al. found the effect of education on health slightly decreases across successive cohorts. After the economic reform, China experienced a remarkable economic growth which improved people’s life quality and result in significant income gap. With the income gap enlarged, opportunities for people to access healthcare were also inequality. By contrast, the income gap in health trajectories diverges for earlier cohorts but converges for most recent cohorts [26]. And in the beginning of 21 century, urban residents have an advantage of 5 years in the life expectancy than rural residents as a result of inequity in acquisition of health care [27]. Li found that the urban-rural disparities of ADL and psychological wellbeing trajectories are decreased with the cohort turning younger [28]. One Japanese research found that BMI among older Japanese with higher education level was lower and it declined linearly at a faster rate over time [29]. Similarity, one Chinese research also proved that the association between education level and self-rated health has positive correlation from older cohort to younger [30]. In this way, trajectories of chronic diseases DALYs may also show SES disparities and the association between SES and DALYs may be interacted by age and cohort.

Moreover, gender disparity of health is also observed in former studies. In most countries, male life expectancy is lower than female life expectancy [31]. Crimmins et al. examined measures of ability to perform ADL and IADL functioning in 13 countries and found that the likelihood of having difficulties in carrying out daily activities and functioning problems was about 2-fold higher for women around the world [32]. For the chronic diseases, though men more likely to have heart disease, stroke, and diabetes, whereas women are more likely to have arthritis and depression [33, 34], the overall gender variations on chronic diseases are not significant [35]. however, the health disparities caused by SES and birth cohort usually show gender differences [36, 37]. This suggest DALYs of chronic diseases may also not show gender disparity but the SES disparity will be different for male and female.

Based on above, there are 3 hypothesizes in this study. Hypothesizes 1: similar to self-rated health, ADL and mental health, trajectory of DALYs of chronic diseases also have cohort disparity due to different life course. Hypothesizes 2: besides the cohort disparity, there exist SES disparities on the trajectories of chronic diseases DALYs and they will be interacted by the cohort. Hypothesizes 3: there are not gender disparity for the DALYs but the SES and cohort disparities will show gender disparity. In order to examined these hypothesizes, this study measured the DALYs for 13 chronic diseases of middle-aged and elderly people by adopting data from 4 waves of China Health and Retirement Longitudinal Study (CHARLS).



Data comes from China Health and Retirement Longitudinal Study (CHARLS) conducted in 2011–2018 ( This study adopted a four-stage, stratified, cluster sampling method to enroll province-dwelling residents from 450 villages and 150 counties in 28 provinces in China. Information was gathered using face-to-face computer-assisted personal interview (CAPI). Aims to provide comprehensive and quality data on the demographic background, family characteristics, health behavior and status, and retirement information of the middle- and old- aged residents in China. This study provides strong data support for analyzing the aging of China’s population. The age range of the sample is defined as 45–90 years old. Our sample consists of 15,101 individuals in 2011, 14,307 in 2013, 13,320 in 2015, 13,310 in 2018. Among them, individuals died by the end of 2018, and loss due to follow-up ranges from 0.07–6.8%. Thus, the sample size for analysis was 55,740 observations (15,062 participants), whose follow-up ranged between 2 and 4 waves. In this study, the DALYs measured 2 waves in 1782 participants, 3 waves in 794 participants, 4 waves in 12,525 participants. More information about the sampling framework can be seen in Fig. 1.

Fig. 1
figure 1

Flow diagram of the sample selection, the China Health and Retirement Longitudinal Study (2011–2018)


DALYs is the lost years of healthy life that caused by diseases and injures which was calculated as the sum of years of life lost (YLLs) and years lived with disability (YLDs). YLLs is the measure for the expected life years lost due to early death and YLDs is the measure for the healthy life lost living with disability and diseases. Disability weight is a critical value for the calculation of YLDs which is a measure of the level of disability of particular health state and diseases, and its values lies between 0 (nearly full health) and 1 (nearly death). In 2012, Global Burden of Disease Study 2010 (GBD 2010) published the measurement of the disability weights of 220 diseases in 195 countries or regions [38]. Then Global Burden of Disease Study 2013 (GBD 2013) further measured disability weights of 235 diseases based on the GBD 2010 [39], which is a large sample, multi-country, and multi-cultural background analysis. Thus, it has the most authoritative and universal results. In this study, valuation of disability weights for chronic diseases is based on the results of GBD 2013. In CHARLS, 14 chronic diseases were referred in the questionnaire, which are hypertension, hyperlipidemia, diabetes, cancer, chronic lung disease, hepatic disease, heart disease, stroke, kidney disease, gastric diseases, emotional and mental illness, memory related diseases, Arthritis/ rheumatism and asthma. According to Chronic Disease Death MICA-ICD-10 Codes, 13 chronic diseases are chosen in this study. Table 1 shows the valuation rules of disability weights for the 13 chronic diseases. Here are notes for the valuation rules in Table 1: (1) Wang et al. calculated the disability weight of hypertension in China based on a large-scale data from many sources [40]. (2) In 2004, WHO measured the disability weight of diabetes as 0.015 with 95%CI of 0.012 to 0.018 [41]. (3) Hepatic diseases usually include viral hepatitis, chronic hepatic diseases and hepatic cancer. In this study, we used the mean value of decompensated liver cirrhosis (the lower limit value) and viral hepatitis to measure the disability weight [42]. In this article, we regarded the YLDs as DALYs. It was because chronic diseases are always non-fatal diseases and, in our study, few participants were died during the surveys, which means the YLDs were approximately equal to the DALYs.

Table 1 Disability weights of 13 chronic diseases

Independent variable

Birth cohort

Birth cohort represented the social set to some degree. In this study, birth cohort was used to reflect the variant life course. In order to distinguish the life course, this study established the 1942–1946, 1947–1953, 1954–1959 and 1960–1964 these four birth cohorts which had distinctive life course, detail division basis on supplementary material Table S1. For a convenience of analysis, birth cohort was regarded as continuous variable which were valued as 1 to 4 for 1942–1947, 1948–1953, 1954–1959 and 1960–1964.

Socioeconomic status

In this study, we use three different measures of SES: educational level, resident place, and per capita family income. To simplify the interpretation, we use three dichotomous variables for education level in the baseline survey. We operationalize educational level as those with illiterate (=1), those with elementary school (=2), those with more than middle school (=3). Since the research subjects are all middle-aged and elderly, the educational level will not change significantly, the educational level of this study mainly used the educational level of the respondents in the baseline survey. We operationalize urban and rural residency as those who living in urban (=0), those who living in rural (=1). The income variable is the per capita income of the respondent’s household, and the logarithm was used to avoid the influence of extreme values.

Control variable

In order to analyze the cohort effect and the influence of socioeconomic status more accurately, other characteristics may affect the middle-aged and elderly health should to be controlled. The CHARLS provides information on individual characteristics variables, including gender, and marital status. Since chronic diseases are also significantly related to the status of receiving medical services, to control for potential medical service variation and health behaviors, we controlled the effects of the inpatient services expenditures and outpatient services expenditures, health behaviors (drinking and smoking), and died/loss to follow-up with dummy variables, using interquartile range to determine the second quartile, expenditures<the second quartile (=0), expenditures > = the second quartile (=1), drinking (=1), not drinking(=0), smoking(=1), not smoking (=0), died/loss to follow-up(=1). At the same time, the self-expense ratio in outpatient and inpatient services were also included as a control variable in this study. Table 1 presents baseline characteristics of the population-based sample and analysis sample.

Statistical analysis

YLDs of 13 chronic diseases are calculated firstly.

$$ {YLDs}_j={W}_j\ast {T}_j $$
$$ {YLDs}_{all}=\sum \limits_{j=1}^{13}{YLDs}_j(2) $$

In Eq. (1), Wj is the disability weight of jth chronic disease. Tj is the years having the disease which can be calculated.

Then the Hierarchical growth curve model is used to analyze the longitudinal data. This modeling reveals significant cohort variations in the age trajectories of health [26]. Most researches illustrated that hierarchical linear model (HLM) or growth curve models can be used to test for cohort differences in age trajectories [43, 44]. And our study figured out the fitting indicators of models (the BIC, AIC, −2LL), found that the quadratic curve fits DALYs change better than simple linear or other curves (See supplementary material Table S2). The hierarchical growth curve model constructed in this study has two levels. The level-1 is repeated measurements within the individual. In this article, it corresponds to the DALYs tracking measurement values of a middle-aged and elderly at different ages. These measurements are nested in the data structure of different middle-aged and elderly in the level-2:


$$ {y}_{ij}={\beta}_{0j}+{\beta}_{1j}\ast {age}_{ij}+{\beta}_{2j}\ast age\hat{\mkern6mu} {2}_{ij}+{e}_{ij} $$

In this model, j represents the individuals from 1, , N samples. yij represents the DALYs of individual j at age i. We center the age variable around, the median age of the cohort to which person i belongs, which can eliminate confounding of age and cohort variables. Age represents the median age i of individual j. Age^2 is the median age square i of individual j. A quadratic term could explain the potential nonlinear effects of age growth. β0j represents the initial value of DALYs at the median age. Other control variables of are included in the outcome parameter model; β1j and β2j represent the slopes of individual DALYs with the measured median age, and eij represents the residual error of an individual j at median age i.


Intercept parameter

$$ {\beta}_{0j}={r}_{00}+{r}_{01}\ast {cohort}_j+{u}_{0j} $$

Slope parameter

$$ {\beta}_{1j}={r}_{10}+{r}_{11}\ast {cohort}_j+{r}_{12}\ast {SES}_j+{r}_{13}\ast {cohort}_j\ast {SES}_j+{u}_{1j} $$
$$ {\beta}_{2j}={r}_{20}+{r}_{21}\ast {cohort}_j+{r}_{22}\ast {SES}_j+{r}_{23}\ast {cohort}_j\ast {SES}_j+{u}_{2j} $$

The aim of the level-2 analysis is to research heterogeneity in change across individuals and to determine the association between predictors (SES and cohort) and the shape of each person’s growth trajectory. β0j called fixed-effect model parameter, represents the influence of cohort characteristics on the intercept in this modeling. The r01 represents the interaction effect between cohort and median age. r10r11, r12 and r13 are coefficients for the parameters of cohort, socioeconomic status, socioeconomic status × cohort on the slope of DALYs, means the interaction effects of these variables with median age. r20r21, r22 and r23 represent the parameters of cohort, socioeconomic status, socioeconomic status × cohort on the slope of DALYs, means the interaction effects of these variables with the square of median age. u0ju1j and u2j are random effects of intercept and slope parameters. Other control variables of individual characteristics that do not change with age or time, such as gender, education level, etc., are placed in the level-2 model. Because people who stay in the sample may be healthier, vulnerable groups are more likely to die early, and the impact of socioeconomic status on health may be overestimated. A simple but effective solution, which is to classify the loss types directly in the level-2 model. The specific approach is entering the dummy variables of death and loss to follow-up (=1) [26], we account for the possibility that those who lost responders due to death or non-response will have high DALYS than survivors with complete data. In addition, for the absence of independent variables, we use multiple imputation methods. All the model were estimated using SAS 9.4. The statistical significance was set at α = 0.05.


Table 2 presents the descriptive statistics of dependent and independent variables tabulated by cohorts. The participants for male and female were 7146 (47.31%) and 7949 (52.66%), respectively. Chronic diseases DALYs for the four cohorts all showed an increasing trend from 2011 to 2018. Participants living in the rural (80.01%) are more than those living in urban (19.99%). With cohorts turning younger, illiterate rate are decreasing and the house income is rising.

Table 2 Descriptive Statistics of All Variables in the Analyses

Table 3 presents findings from the estimation of growth curve models. We conducted analysis of cohort, SES indicators (resident place, income and education level) and the interactions of SES-by-cohort from Model 1 to Model 3. Model 1 strongly support Hypothesis 1: there exist significant cohort differentials in trajectories of middle-aged and elderly people’s chronic diseases DALYs. Model 1 shows the age trajectory of DALYs after controlling the cohort effect, which has a mean of 1.55 (95%CI: 1.086, 2.014) years and increases as a rate of 0.49 (95%CI: 0.469, 0.511) years per year of age, slowing at a rate of 0.005 with age, thereby exhibiting a quadratic pattern. Results also shows there are strongly significant cohort variations in the intercept or mean level of DALYs. Respondents in younger cohorts has a 0.232 (95%CI: − 0.305, − 0.16) years less than older cohorts for the DALYs intercept and its growth with age show less rapid ((the cohort-by-age interaction is − 0.143 with 95% CI: − 0.152, − 0.134). For a better view of these trajectories, based on estimates of Model 1 we plotted Fig. 2 in below. From Fig. 2, it can be found that older cohorts have a faster growth of DALYs with age, however, chronic diseases of older cohorts seem appear in a later age than younger cohorts.

Table 3 Cohort and socioeconomic disparities of chronic diseases DALYs
Fig. 2
figure 2

Cohort disparity of chronic diseases DALYs trajectories

Model 2 and 3 show findings support for Hypothesis 2: SES would affect the growth trajectories of DALYs and it was interacted by cohort effect. Model 2 tests the SES indicators disparities of DALYs age trajectories and find there exist significant SES gradients in the intercept or mean level of DALYs. Relative to the most disadvantaged (individuals whose education level were illiterate and elementary school), respondents with middle school and higher education levels have less DALYs by 0.299 (95%CI:0.092,0.505) and 0.342 (95%CI:0.161,0.522) years and its growth trend with age was slower (the age-by-education level interaction is 0.035 with 95%CI: 0.021, 0.049 and 0.028 with 95%CI: 0.015, 0.04). Compared with individuals living in rural area, respondents living in urban have a lower DALYs by 0.666 (95%CI: − 0.867, − 0.464) years and its growth with age is also slower (the age-by-resident place interaction is − 0.037 with 95%CI: − 0.051, − 0.024). Consistent with resident place and education level, respondents with higher income have lower DALYs by 0.235 (95%CI: − 0.363, − 0.107) years by one unit, and its age growth trajectory still show less rapid (the age-by-income interaction is − 0.021 with 95%CI: − 0.029, − 0.012). Model 3 added interactions of SES indicators by cohort to explore the SES differentials across cohorts. Model 3 reports the significant age-cohort-SES indicators interactions which suggest the DALYs SES disparities have differentials across cohorts. The age-by-resident place-by-cohort interaction is 0.039 (95%CI: 0.018,0.059). This indicate the DALYs disparities of resident place are narrowed with cohort turning younger. Similarly, we find the education level and income disparities of DALYs are also narrowed with cohort turning younger (the age-by-education level-by-cohort interaction is − 0.041 and − 0.034 with 95%CI: − 0.062, − 0.021 and − 0.052, − 0.016; the age-by-income-by-cohort interaction is 0.039 with 95%CI: 0.025,0.052). For a better interpret of these trajectories, we plotted age trajectories of DALYs by resident place in Fig. 3. From Fig. 3, we can find DALYs growth trajectory of urban respondents are slower than those live in rural and resident place differentials are narrowed in younger cohorts.

Fig. 3
figure 3

Resident place disparity of chronic diseases DALYs

Table 4, which mainly support Hypothesis 3, presents the cohort and SES disparities of DALYs across gender by conducting analyses for male and female separately. From the results, it is found to be differences in growth trajectories of DALYs for respondents with different cohort and SES indicators across gender. Specifically, disparities of DALYs growth trajectories for male caused by education level and resident place will not be narrowed with cohort turning younger (the age-by-education level-by-cohort interaction and age-by-resident place interaction are not significant). On the contrary, these disparities are narrowed for female with cohort turning younger (the cohort-by-age-by-education level interaction are − 0.054 with 95%CI: − 0.102,0.006 and − 0.042 with 95%CI: − 0.007, − 0.015 and the age-by-resident place-by-cohort interaction are 0.075 with 95%CI: 0.041,0.11).

Table 4 Cohort and SES disparity of chronic diseases DALYs across gender


The objective of this study is to explore the growth trajectory of DALYs of chronic diseases for middle-aged and elderly people and its disparities across cohort, SES (resident place, income and education level) and gender. We find there are significant cohort and SES disparities in the growth trajectories and the SES and cohort disparities will show differentials across different gender.

In this study, we calculate the chronic diseases DALYs by valuing the disability weights from GBD study 2013. This is a creative application for the GBD study which is firstly used in the exploration of early psychological risks effects on health [45], and the results of this study further proved the validity of this methods.

Our findings suggest that chronic diseases DALYs of earlier cohorts show more rapid growth trend with age than the recent cohorts but the diseases will be developed in a later age. This indicates cohort effect is also significant for chronic diseases DALYs which is consistent with former researches that focus on elderly health. Cohort represents the life course that individuals live through (The life course of four cohorts in this study were presents in the appendix Table S1). In China, for the olds of earlier cohorts, they often suffered more negative events in their life course. For example, individuals from cohort 1942–1948 suffered wars (1937–1949) in their children stage and encountered the great famine (1959–1961) in the youth and then live 10 years through the political turmoil (the Great Cultural Revolution, 1966–1976). However, the youngest cohort in this study (cohort 1960–1965) only suffered the political turmoil in their child stage. According to the accumulation of risk model, effects accumulate over the life course, as results health damage will increase with the duration and/or number of detrimental exposures [46]. In this way, earlier cohorts have a higher DALYs than recent cohorts. Compared with the earlier, recent cohorts are generally richer and own a higher life quality, which are often more likely to have unhealthy diets, to be obese, to smoke and drink more, and to be sedentary [47, 48]. This explained reasons why the chronic diseases are developed earlier than individuals in older cohorts.

In addition, this study also finds DALYs of cohort 1954–1959 have lower growth trend than cohort 1947–1953. Individuals from these cohorts have similarly detrimental exposures which indicates they should also have the similar growth trend. This differential can be explained by the theory of critical period model which stressed the critical role of timing of one exposure [49]. Difference between these cohorts is the time they suffered the Great Cultural Revolution. For individuals in cohort 1948–1953, the Great Cultural Revolution happened in their 13 to 18 years old and lasted to their 23 to 28 years old. This period was a critical period for their life development. However, for those in cohort 1954–1959, this negative event happened in their late childhood, and last to their 17 to 22. Though influenced by this event, most of people caught up with the series reforms after the political turmoil, which reduced the negative effect to some degree [50].

Besides cohort disparity, our study finds there exists resident place, education level and income differentials for the age trajectory of chronic diseases DALYs and these disparities are enlarged/ diverged with age. This is consistent with existing researches and supported by the cumulative disadvantage/advantage theory: the positive effect of higher SES is cumulative increased by influencing recourses such as access to healthcare, health behavior and social support [51]. At the same time, it is also found these SES disparities are narrowed/converged with cohort turning younger. This can be explained by the efforts that government made to improve social medical insurance. In the last 10 years, China’s social medical insurance cover rate has been greatly improved so that individuals from younger cohort are mostly benefited [52]. On the other hand, with the urbanization, income and education level improved, SES disparities among recent cohorts are gradually narrowed [53].

Besides, it is also found resident place and education level disparities of DALYs across cohort have gender differentials. For younger female, in the last decades, their education level is improved evidently, and the government greatly improved the maternal health care especially in the rural area [54]. As results, DALYs disparities of these SES indicators are significantly narrowed across cohort. This also indicates improvements of education level and health care will have more benefit for female.

This study has several limitations. Firstly, DALYs is the sum of YLLs and YLDs, but because chronic diseases are always chronic non-fatal diseases of which the DALYs are mainly is the YLDs. In this article, we regarded the YLDs of elderly’s chronic diseases as their DALYs. Secondly, data used in this study was from 2011 to 2018, but the disability weights were from the GBD study 2013. Though the disability weights have a high universality, but it would be better to have the weight values of 2011, 2015 and 2018 which were lacked in real. Then, if mortality and non-response are significantly associated with higher DALYS, non-random selection may occur, resulting in a biased sample estimate of the trajectory. Even though we used the dummy variables of death and loss to follow-up. But it will inevitably be led to a distortion of the authenticity of DALYS calculation results. Finally, in order to examine the cohort effect from the life course perspective, the cohort in our study was limited from 1942 to 1965, which are not the full data.

Despite these limitations, this study has some advantages. Firstly, the longitudinal design of CHARLS allowed for more accumulation of information on age than single cross-sectional study. Secondly, this study is the first to measure 13 chronic diseases by calculating their DALYs, which can provide more precise and comprehensive understand of the chronic diseases’ growth trajectory in old stage.


(1) DALYs of chronic diseases for middle-aged and elderly people present a general growing trend with age. Individuals having more negative event exposures in their life course show a more rapid growth trend and the timing of turmoil will also affect the DALYs trajectory. The effect of life course on chronic diseases is accumulated with exposures and its exposure in critical period is also forceful.

(2) Chronic diseases of individuals in recent cohorts are developed earlier than early cohorts. In developing countries like China, individuals in recent cohort have higher SES, but tend to have unhealthy life behaviors.

(3) Living in urban, owning higher education level and income will decrease the DALYs of chronic diseases, and advantages brought by these SES indicators will be accumulated with age. Government efforts in improving social medical insurance will lower the SES disparity which greatly benefit individuals from recent cohorts. Thus, the SES disparities show a narrowing trend with cohort turning younger.

(4) Compared with male, improvements of education level, health care and life quality will more benefit the female. The SES disparities of resident place and education level is narrowed with cohort turning younger for female. However, there no such trend for male. Governments should pay more attentions for male’s chronic diseases.

Availability of data and materials

CHARLS data are available at (requiring a simple application).



Disabled Adjusted Life Years


  1. Wang LM, Chen ZH, Zhang M, Zhao ZP, Huang ZJ, Zhang X, et al. Study of the prevalence and disease burden of chronic disease in the elderly in China. Zhonghua Liu Xing Bing Xue Za Zhi. 2019;40(3):277–83.

    Article  CAS  PubMed  Google Scholar 

  2. Fang EF, Xie C, Schenkel JA, Wu C, Long Q, Cui H, et al. A research agenda for ageing in China in the 21st century (2nd edition): focusing on basic and translational research, long-term care, policy and social networks. Ageing Res Rev. 2020;64:101174.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Ordunez P, Nieto FJ, Martinez R, Soliz P, Giraldo GP, Mott SA, et al. Chronic kidney disease mortality trends in selected Central America countries, 1997-2013: clues to an epidemic of chronic interstitial nephritis of agricultural communities. J Epidemiol Community Health. 2018;72(4):280–6.

    Article  PubMed  Google Scholar 

  4. Kontopantelis E, Mamas MA, van Marwijk H, Ryan AM, Bower P, Guthrie B, et al. Chronic morbidity, deprivation and primary medical care spending in England in 2015-16: a cross-sectional spatial analysis. BMC Med. 2018;16(1):19.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Geiger BB. Has working-age morbidity been declining? Changes over time in survey measures of general health, chronic diseases, symptoms and biomarkers in England 1994-2014. BMJ Open. 2020;10(3):e032378.

    Article  PubMed  PubMed Central  Google Scholar 

  6. González-Henares MA, Clua-Espuny JL, Lorman-Carbo B, Fernández-Saez J, Queralt-Tomas L, Muria-Subirats E, et al. Risk of Long-term mortality for complex chronic patients with Intracerebral hemorrhage: a population-based e-cohort observational study. Adv Ther. 2020;37(2):833–46.

    Article  CAS  PubMed  Google Scholar 

  7. Moreira NF, Sichieri R, Reichenheim ME, Oliveira AS, Veiga GV. The associations of BMI trajectory and excessive weight gain with demographic and socio-economic factors: the adolescent nutritional assessment longitudinal study cohort. Br J Nutr. 2015;114(12):2032–8.

    Article  CAS  PubMed  Google Scholar 

  8. Wang K, Chen X, Gerke TA, Bird VY, Ghayee HK, Prosperi M. BMI trajectories and risk of overall and grade-specific prostate cancer: an observational cohort study among men seen for prostatic conditions. Cancer Med. 2018;7(10):5272–80.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Stuart B, Panico L. Early-childhood BMI trajectories: evidence from a prospective, nationally representative British cohort study. Nutr Diabetes. 2016;6(3):e198.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Stenholm S, Virtanen M, Pentti J, Oksanen T, Kivimäki M, Vahtera J. Trajectories of self-rated health before and after retirement: evidence from two cohort studies. Occup Environ Med. 2020;77(2):70–6.

    Article  PubMed  Google Scholar 

  11. Chin WY, Choi EP, Wan EY. Trajectory pathways for depressive symptoms and their associated factors in a Chinese primary care cohort by growth mixture Modelling. PLoS One. 2016;11(2):e0147775.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. Dowd JB, Zajacova A. Does the predictive power of self-rated health for subsequent mortality risk vary by socioeconomic status in the US? Int J Epidemiol. 2007;36(6):1214–21.

    Article  PubMed  Google Scholar 

  13. Debiasi E, Dribe M. SES inequalities in cause-specific adult mortality: a study of the long-term trends using longitudinal individual data for Sweden (1813-2014). Eur J Epidemiol. 2020;35(11):1043–56.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Regidor E, Guallar-Castillón P, Gutiérrez-Fisac JL, Banegas JR, Rodríguez-Artalejo F. Socioeconomic variation in the magnitude of the association between self-rated health and mortality. Ann Epidemiol. 2010;20(5):395–400.

    Article  PubMed  Google Scholar 

  15. Schrage B, Lund LH, Benson L, Stolfo D, Ohlsson A, Westerling R, et al. Lower socioeconomic status predicts higher mortality and morbidity in patients with heart failure. Heart. 2021;107(3):229–36.

    Article  PubMed  Google Scholar 

  16. Goldberg S, Fruchter E, Davidson M, Reichenberg A, Yoffe R, Weiser M. The relationship between risk of hospitalization for schizophrenia, SES, and cognitive functioning. Schizophr Bull. 2011;37(4):664–70.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Zhang D, Matthews CE, Powell-Wiley TM, Xiao Q. Ten-year change in neighborhood socioeconomic status and colorectal cancer. Cancer. 2019;125(4):610–7.

    Article  PubMed  Google Scholar 

  18. Weinberg D, Stevens G, Duinhof EL, Finkenauer C. Adolescent Socioeconomic Status and Mental Health Inequalities in the Netherlands, 2001-2017. Int J Environ Res Public Health. 2019;16(19):3605.

    Article  PubMed Central  Google Scholar 

  19. Meyer OL, Castro-Schilo L, Aguilar-Gaxiola S. Determinants of mental health and self-rated health: a model of socioeconomic status, neighborhood safety, and physical activity. Am J Public Health. 2014;104(9):1734–41.

    Article  PubMed  PubMed Central  Google Scholar 

  20. Evans-Lacko S, Aguilar-Gaxiola S, Al-Hamzawi A, Alonso J, Benjet C, Bruffaerts R, et al. Socio-economic variations in the mental health treatment gap for people with anxiety, mood, and substance use disorders: results from the WHO world mental health (WMH) surveys. Psychol Med. 2018;48(9):1560–71.

    Article  CAS  PubMed  Google Scholar 

  21. Rossow I, Amundsen EJ, Samuelsen SO. Socio-economic differences in all-cause mortality in people with alcohol use disorder: a prospective cohort study. Addiction. 2021;116(1):53–9.

    Article  PubMed  Google Scholar 

  22. Avan A, Digaleh H, Di Napoli M, Stranges S, Behrouz R, Shojaeianbabaei G, et al. Socioeconomic status and stroke incidence, prevalence, mortality, and worldwide burden: an ecological analysis from the Global burden of disease study 2017. BMC Med. 2019;17(1):191.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Signorello LB, Cohen SS, Williams DR, Munro HM, Hargreaves MK, Blot WJ. Socioeconomic status, race, and mortality: a prospective cohort study. Am J Public Health. 2014;104(12):e98–e107.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Ng R, Sutradhar R, Yao Z, Wodchis WP, Rosella LC. Smoking, drinking, diet and physical activity-modifiable lifestyle risk factors and their associations with age to first chronic disease. Int J Epidemiol. 2020;49(1):113–30.

    Article  PubMed  Google Scholar 

  25. Zhu B, Ye Y. Gender disparities in the education gradient in self-reported health across birth cohorts in China. BMC Public Health. 2020;20(1):375.

    Article  PubMed  PubMed Central  Google Scholar 

  26. Chen F, Yang Y, Liu G. Social change and socioeconomic disparities in health over the life course in China: a cohort analysis. Am Sociol Rev. 2010;75(1):126–50.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Wang F, Mason A. Population ageing: challenges, opportunities, and institutions. UK: Transition & Challenge; 2007.

    Google Scholar 

  28. Li T, Zhang Y. Growth curve of elderly health indicators under birth cohort effect and its urban-rural difference. Population Res. 2014;38:18–35.

    Google Scholar 

  29. Murayama H, Liang J, Bennett JM, Shaw BA, Botoseneanu A, Kobayashi E, et al. Socioeconomic status and the trajectory of body mass index among older Japanese: a Nationwide cohort study of 1987-2006. J Gerontol B Psychol Sci Soc Sci. 2016;71(2):378–88.

    Article  PubMed  Google Scholar 

  30. Zheng L, Zeng X. The Urban-Rural and Gender Differences in Educational Health Returns: An Analysis Based on the Effect of Generation and Age. J East China Univ Sci Technol. 2018;33:54–65.

    Google Scholar 

  31. Barford A, Dorling D, Davey Smith G, Shaw M. Life expectancy: women now on top everywhere. Bmj. 2006;332(7545):808.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Crimmins EM, Kim JK, Solé-Auró A. Gender differences in health: results from SHARE, ELSA and HRS. Eur J Pub Health. 2011;21(1):81–91.

    Article  Google Scholar 

  33. Kingston A, Davies K, Collerton J, Robinson L, Duncan R, Bond J, et al. The contribution of diseases to the male-female disability-survival paradox in the very old: results from the Newcastle 85+ study. PLoS One. 2014;9(2):e88016.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  34. Sobhani K, Nieves Castro DK, Fu Q, Gottlieb RA, Van Eyk JE. Noel Bairey Merz C: sex differences in ischemic heart disease and heart failure biomarkers. Biol Sex Differ. 2018;9(1):43.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Crimmins EM, Shim H, Zhang YS, Kim JK. Differences between men and women in mortality and the health dimensions of the morbidity process. Clin Chem. 2019;65(1):135–45.

    Article  CAS  PubMed  Google Scholar 

  36. Lynch SM. Cohort and life-course patterns in the relationship between education and health: a hierarchical approach. Demography. 2003;40(2):309–31.

    Article  PubMed  Google Scholar 

  37. Ross CE, Masters RK, Hummer RA. Education and the gender gaps in health and mortality. Demography. 2012;49(4):1157–83.

    Article  PubMed  Google Scholar 

  38. Murray CJ, Vos T, Lozano R, Naghavi M, Flaxman AD, Michaud C, et al. Disability-adjusted life years (DALYs) for 291 diseases and injuries in 21 regions, 1990-2010: a systematic analysis for the Global burden of disease study 2010. Lancet. 2012;380(9859):2197–223.

    Article  PubMed  Google Scholar 

  39. Salomon JA, Haagsma JA, Davis A, de Noordhout CM, Polinder S, Havelaar AH, et al. Disability weights for the Global burden of disease 2013 study. Lancet Global Health. 2015;3(11):E712–23.

    Article  PubMed  Google Scholar 

  40. Wang J, Zhang Q, Jin S. Analysis of disease burden of hypertension and diabetes in China. Public Health Prev Med. 2007;03:27–30.

    Google Scholar 

  41. World Health Organization. Global burden of disease 2004 update: disability weights for diseases and conditions Retrieved Mar 29 2021, from:

  42. Wong MCS, Huang JLW, George J, Huang J, Leung C, Eslam M, et al. The changing epidemiology of liver diseases in the Asia-Pacific region. Nat Rev Gastroenterol Hepatol. 2019;16(1):57–73.

    Article  PubMed  Google Scholar 

  43. Raudenbush SW, Bryk AS. Hierarchical linear models: applications and data analysis methods. Thousand Oaks: Sage; 2002.

    Google Scholar 

  44. Miyazaki Y, Raudenbush SW. Tests for linkage of multiple cohorts in an accelerated longitudinal design. Psychol Methods. 2000;5(1):44–63.

    Article  CAS  PubMed  Google Scholar 

  45. Gao M. The effect of early psychosocial risk on health: based on the data from the China Health and Retirement Longitudinal Survey. China Soc Sci. 2020; 296(09):93-116+206.

  46. Lynch J, Smith GD. A life course approach to chronic disease epidemiology. Annu Rev Public Health. 2005;26(1):1–35.

    Article  PubMed  Google Scholar 

  47. Kim S, Symons M, Popkin BM. Contrasting socioeconomic profiles related to healthier lifestyles in China and the United States. Am J Epidemiol. 2004;159(2):184–91.

    Article  PubMed  Google Scholar 

  48. Du S, Lu B, Zhai F, Popkin BM. A new stage of the nutrition transition in China. Public Health Nutr. 2002;5(1a):169–74.

    Article  PubMed  Google Scholar 

  49. Hertzman C, Power C, Matthews S, Manor O. Using an interactive framework of society and lifecourse to explain self-rated health in early adulthood. Soc Sci Med. 2001;53(12):1575–85.

    Article  CAS  PubMed  Google Scholar 

  50. Liu Y. A Study on Childhood Migration Experience and Early Adult Life Opportunity from the Perspective of Life Course. Population Res. 2013;37:93–101

    Google Scholar 

  51. Dannefer D. Systemic and reflexive: foundations of cumulative dis/advantage and life-course processes. J Gerontol B Psychol Sci Soc Sci. 2020;75(6):1249–63.

    Article  PubMed  Google Scholar 

  52. Pan J, Tian S, Zhou Q, Han W. Benefit distribution of social health insurance: evidence from china's urban resident basic medical insurance. Health Policy Plan. 2016;31(7):853–9.

    Article  PubMed  Google Scholar 

  53. You X, Zhang Y, Zeng J, Wang C, Sun H, Ma Q, et al. Disparity of the Chinese elderly's health-related quality of life between urban and rural areas: a mediation analysis. BMJ Open. 2019;9(1):e024080.

    Article  PubMed  PubMed Central  Google Scholar 

  54. Wu Y, Zhou H, Wang Q, Cao M, Medina A, Rozelle S. Use of maternal health services among women in the ethnic rural areas of western China. BMC Health Serv Res. 2019;19(1):179.

    Article  PubMed  PubMed Central  Google Scholar 

Download references


We thank the China Health and Retirement Longitudinal Study (CHARLS) team for providing data and training in using the datasets. We thank the students who participated in the survey for their cooperation. We thank all volunteers and staff involved in this research.


National Planning Office of Philosophy and Social Science, Award number: 21ZDA104, Recipient: Peigang Wang.

Author information

Authors and Affiliations



GMZ, FT are the joint first authors, who contributed equally to the manuscript. GMZ and FT conceived and designed the study. GMZ searched literature, GMZ and FT analyzed data. GMZ and FT wrote the first draft of the manuscript. GMZ, FT, JL, and PGW modified the manuscript and language, GMZ and FT reviewed the manuscript. All authors critically revised the manuscript and approved the final version.

Corresponding author

Correspondence to Peigang Wang.

Ethics declarations

Ethics approval and consent to participate

Data used in this article comes from China Health and Retirement Longitudinal Study (CHARLS), which was an open data and own ethics approval and consent to participate.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests. No conflict of interest exits in the submission of this manuscript, and manuscript is approved by all authors for publication. This work was original research that has not been published previously, and not under consideration for publication elsewhere. The authors listed have approved the manuscript that is enclosed.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and Permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhang, G., Tang, F., Liang, J. et al. Trajectories of middle-aged and elderly people’s chronic diseases Disability Adjusted Life Years (DALYs): cohort, socio-economic status and gender disparities. Int J Equity Health 20, 179 (2021).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI:


  • DALYs
  • Chronic diseases
  • Life course
  • SES
  • Gender