With the declining fertility rate and the development of modern medical care, the aging process in China accelerates (1). According to the Seventh National Population Census of the People’s Republic of China, in 2020, the people aged 60 and above, 65 and above accounted for 18.70 and 13.50% of the total Chinese population, respectively (2). It is projected that by 2050, the proportion of people aged 65 and above will increase to 26.9% of the total Chinese population (3). Meanwhile, as the prevalence of chronic diseases among the older adults continues to grow, the associated burden on public health and socioeconomic development becomes more severe (4). Hence, cost-effective health indicators for the older adults are expected to help public health institutions identify high-risk groups and allocate resources effectively to improve health management in an aging society.

In recent years, simplified indicators of health status have gained attention from researchers. One such indicator is Self-Rated Health (SRH), which measures respondents’ own evaluation of their health status, and encompasses physical health, mental health, and social participation simultaneously (5, 6). Previous studies have confirmed the reliability of SRH in predicting overall health (7), and the World Health Organization (WHO) has recommended SRH as a cost-effective and reliable surrogate indicator of health status (8). However, SRH may exhibit disparities among individuals due to a set of cognitive factors, such as different understanding of health. Interviewer-Rated Health (IRH) addresses this limitation by enabling interviewers to evaluate respondent’s health status during interviews (9). Interviewers often have a more comprehensive and objective perception of health than respondents themselves, and thus IRH can provide supplemental information to SRH, such as respondents’ living environments and lifestyle habits (9). However, these subjective health indicators may lack specificity and pose challenges in ensuring the accuracy of information, particularly when they are used to determine eligibility for further care resources (10). In contrast to subjective health, objective health refers to the health status of respondents characterized by objectively measured biochemical or functional indicators. Recently, the Chinese multimorbidity-weighted index (CMWI) has been proposed as an objective health evaluation index that can quantify the burden of multimorbidity among Chinese older adults through routinely physical examinations at the community health care center (11).

Previous studies have illustrated that poor subjective and objective health were associated with an increased risk of mortality in older adults (12–18). A study involving 373,761 participants from the UK Biobank investigated the associations of different combinations of SRH and objective health status with all-cause mortality, and found that individuals with concordant favorable SRH and health status had the lowest hazard of mortality, while those with concordant unfavorable SRH and health status had the highest hazard of mortality, and those with discordant SRH and health status had intermediate hazards of mortality (19). However, studies also indicated that health status is dynamic and can change over a long follow-up period (20). Some studies have explored the associations of the change and trajectory of SRH with mortality, but the findings have been inconsistent (21–24). For instance, the Lifelines Cohort Study suggested that older adults individuals with a stable poor SRH trajectory often had unfavorable health status (21). While another study with 3,129 participants indicated no significant associations between SRH trajectories and mortality (24). Besides, it remains unclear whether changes and trajectories of IRH and CMWI over time are associated with mortality.

To fill this research gap, we aimed to investigate (1) the association of baseline SRH, IRH, and CMWI with mortality; (2) the association of changes of SRH, IRH, and CMWI (the values of SRH, IRH, and CMWI in 2014 minus the values in 2008) with mortality; (3) the association of trajectories of SRH, IRH, and CMWI (constructed by the values of SRH, IRH, and CMWI in 2008, 2011, and 2014) with mortality.

Our study found that baseline SRH, IRH, and CMWI were all negatively associated with 10-year mortality in Chinese older adults. Better changes and trajectories of SRH, IRH, and CMWI were also significantly associated with lower 4-year mortality in Chinese older adults.

Previous studies have revealed the negative associations of baseline SRH, IRH, and objective health with mortality, which were in line with our results (11, 16, 29, 32, 33). First, SRH is a subjective assessment based on objective health status (34), therefore serves as a statistical indicator with a condensed overview of physical information (5). Second, compared to objective indicators of health, SRH might capture the burden of subclinical conditions, such as high blood pressure, early stages of diabetes, and preclinical forms of cancer (35). Third, poor SRH sometimes mirrors mental stress, which can lead to elevations in serum proinflammatory cytokines (36–38). Inflammation is associated with symptoms of sickness behavior, such as lethargy, decreased appetite, and behavioral withdrawal, which in turn will have an impact on SRH (39). Additionally, the dysregulation of inflammation leads to the development of illnesses and raises the risk of mortality in older adults (40–42). Consequently, elevated levels of circulating inflammatory markers contribute to both a decrease in SRH and an increase in mortality risk (43).

Some recent studies reported that IRH is a better indicator of mortality than SRH, and investigated the potential mechanisms underlying the association. For instance, using data from CLHLS, Feng et al. found that the association between IRH and mortality remained significant even after adjusting for SRH (7). Similarly, Todd M A and Goldman N used an older adults cohort in Taiwan to compare the predictive power of SRH, IRH, and health ratings from physicians on 5-year mortality, and revealed that IRH was the most accurate predictor (16). Several reasons have been put forward to explain why IRH may be a better predictor of mortality than SRH. First, when evaluating SRH, respondents may underestimate health-related behaviors and functional limitations (44), whereas interviewers who assess IRH can weigh and integrate individual information more effectively (45). Moreover, interviewers may be able to capture unmeasured health information that SRH lacks through on-site observations, such as respondents’ appearance and living situation. Second, in CLHLS, the rating process is done immediately following the interviewer’s completion of data collection, therefore IRH incorporates detailed information about the participants overall health, including their SRH. Third, trained interviewers can maintain a consistent and relatively objective evaluation standard when investigating multiple respondents, avoiding personal biases to some extent. Furthermore, interviewers generally have higher educational attainment than the respondents, which could help them accurately summarize information during the interview and provide a more accurate assessment of the respondent’s health conditions (7, 16, 44). Incorporating both self-rated and interviewer-rated health could provide a more complete understanding of a person’s health status.

Previous studies produced contradictory results of gender differences in the association between SRH and mortality (46). Our results indicate that SRH was associated with mortality in male. In females, however, there was no association between the change of SRH and mortality. These sex differences in the associations between subjective health indicators and mortality may result from the different health self-perceptions, which means the meaning of “health” differs between sexes (46). Also, the covariates may be different between males and females. Besides, it is noteworthy that sex-stratified analyses showed that compared to male, the association between CMWI and mortality was not significant in females, which was consistent with result from Wuorela et al. (47). One explanation could be that although women have poorer health status, they can tolerate more frailty than men. It could also result from biological and social factors (48, 49).

Multimorbidity, defined as the presence of two or more chronic diseases in the same individual, reflects worse health status and is associated with mortality (50). Previous study has shown that CMWI is a superior predictor of mortality among the Chinese older adults compared to other multimorbidity indexes, as its disease weights were tailored to the Chinese middle-aged and older community-dwelling adults (11). When stratified by sex, we found that baseline CMWI, together with its change and trajectory was not associated with subsequent mortality in female. This may be due to several limitations. First, mental function was not considered in the development and validation of CMWI due to limited mental health measures. This may result in information bias, as there are sex-differences in mental health conditions. Second, the collection of information on chronic diseases were largely gathered by self-report, which may lead to reporting bias. Third, the burden of multimorbidity may have been underestimated because the severity of the disease was not taken into account while calculating CMWI. Thus, further studies are needed to investigate the sex-specific factors behind CMWI and explore a more universally applicable indicator of mortality for both males and females.

Previous studies have investigated the association between changes and trajectories in SRH and mortality, but the results have not been consistent. For instance, the Women’s Health and Aging Study suggested that change in SRH was a stronger indicator of mortality than baseline SRH (23). While another study found that SRH trajectories did not have a significant, direct impact on mortality (24). The inconsistent findings might stem from variations in investigation date, sample size, studying setting, and other factors. Our study confirmed significant associations of changes and trajectories in SRH with mortality, and further indicated that the changes and trajectories of IRH and CMWI were also associated with mortality in Chinese older adults. These indicators are dynamic and change over time during the follow-up period (46, 51). Neglecting to assess SRH as a dynamic indicator may lead to a misinterpretation of the association between SRH and mortality (52). One study revealed that baseline SRH was only associated with 4-year mortality, but not with 9-year mortality among old female (53). A research also indicated that poor baseline SRH was associated with higher risk of 10-year mortality, but no longer with 18-year mortality in older male (38). Idler and Benyamini proposed the trajectory hypothesis, which suggested that people incorporated changes in prior health status into their current health evaluations, and that poor SRH indicated assessments of their future decline or death (24, 29). Since the changes and trajectories are originated from different interview points, their evaluation often involve all the information of multiple time points. Besides, changes in other information such as socioeconomics factors, which are not included in the questionnaire can be reflected in the changes and trajectories of indicators (7). Therefore, the changes and trajectories of indicators contain a wider range of information than baseline.

Numerous studies have explored a range of mortality indicators, such as multimorbidity-weighted index and the Elixhauser comorbidity index. However, their widespread use in densely populated developing countries like China is challenging due to their high cost and implementation complexity (11). Besides, previous studies have also shown that the combination of subjective and objective health indicators could provide a simple measure for risk classification (19). Our study highlights the applicability of SRH, IRH, and CMWI, particularly in geriatric health-related studies. Because SRH, IRH, and CMWI are easy to use and cost-effective, it is realistic to popularize them in grassroots communities and rural areas in China, and conducive to the government’s preliminary understanding of the health status of the local older adults. Formulating a more appropriate health promotion plan for the older adults can promote the healthy aging construction of Chinese society.

To the best of our knowledge, this study was the first prospective cohort study to comprehensively explore the association of baseline SRH, IRH, and CMWI, their changes and trajectories with subsequent mortality in Chinese older adults. Moreover, with the national sample, the CLHLS represented community-based older adults from 22 Chinese provinces, ensuring that our findings were reliable and generalizable.

This study had several limitations. First, we merged “poor” and “very poor” SRH due to the limitations of sample size, which may lead to information bias. Also, since we used the continuous forms of SRH and IRH, the results of baseline values and changes of SRH and IRH with mortality might be difficult to interpret. However, the sample sizes in some groups were relatively small when considering SRH and IRH as categorial variables. We have conducted sensitive analyses to investigate the associations of SRH and IRH as categorical variables with 10-year mortality. The results were consistent with the main findings, supporting the robustness of our approach. Second, the number of participants included in the analysis of the trajectory was relatively small, thus, the trajectory of IRH was not smooth enough. However, this limitation did not influence the association between trajectories of indicators and mortality. Further studies are needed in a larger population. Third, some of the covariates, such as physical activity and smoking history, were lacking more detailed data in the questionnaire of CLHLS, which may result in potential residual confounding in the associations. Finally, some potential covariates, such as type of work, social participation, may have been neglected in the adjustment.

SRH, IRH, and CMWI are practical and accurate indicators of mortality in the Chinese older adults. More researches are required to promote the application and generalization of cost-effective indicators in the field of health assessment and monitoring among Chinese older adults.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding authors.

PS designed the study. ZR and SS managed and analyzed the data. SS and SC prepared the first draft. SS, JC, SC, KT, SL, WS, LH, and QY reviewed and edited the manuscript, with comments from PS. All authors involved in revising the manuscript, had full access to the data, and gave final approval of the submitted versions.

Data used for this research was provided by the study entitled “Chinese Longitudinal Healthy Longevity Survey” (CLHLS) managed by the Center for Healthy Aging and Development Studies, Peking University. CLHLS was supported by funds from the U.S. National Institutes on Aging (NIA), China Natural Science Foundation, China Social Science Foundation, and UNFPA. The study sponsors had no role in the study design, analysis, interpretation of data, writing of the report, or in the decision to submit the paper for publication.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.