Frailty is an age-dependent syndrome that diminishes functioning across multiple physiological systems while increases responsiveness to various stressors (1). It is intricately associated with numerous undesirable outcomes, including mortality (2), disability (3), falls (4), fractures (5), low quality of life (6), and increased health-care expenditure and utilization (7). With a rising aging population and a longer life expectancy, older people suffering from frailty are dramatically increasing. Frailty is emerging as a global health burden, posing great challenges to health, aging and healthcare system, particularly, in those low- and middle-income locations (1).

Being a populous country, China has a rapidly growing older population (8). According to the data published by the National Bureau of Statistics of China, in the year 2020, China had 264 million people aged 60 years or older, accounting for 18.7% of the total 1.41 billion people (9). And this population is estimated to be 28% by the year 2040, given the enhanced survival durations and vastly reduced fertility rates (8). The biggest challenge of population aging is frailty (10). A previous research, based on the China Health and Retirement Longitudinal Study (CHARLS), has revealed that frailty is more common at advanced ages, and there were 51.2% prefrail and 7.0% frail adults among those aged ≥60 years in 2011 (11), which greatly burdened the long-term medical care in China. A comprehensive insight into frailty risk factors may provide opportunities to prevent and manage frailty, especially when these risk factors are manageable.

Multiple sociodemographic, lifestyle, clinical, and biological conditions (e.g., low education, physical inactivity, multimorbidity) are correlated with frailty (1, 12). In recent epidemiologic studies, environmental factors like air pollution were demonstrated to enhance frailty risk among older adults (13–17). Nonetheless, current publications mostly focused on outdoor air pollution (14–17). Till date, only one research explored the association between household air pollution (HAP) and frailty (13). Although HAP from solid fuels has declined markedly over the last few years, 36% of the Chinese population still relies on biomass or coal fuels for cooking (18), which, unfortunately, accounts for ~1 million premature deaths in China each year (19). Given the rapidly aging population and the common usage of biomass fuels, assessing the influence of HAP from biomass fuels on frailty is both urgent and necessary.

Herein, we employed the Chinese Longitudinal Healthy Longevity Survey (CLHLS), a longitudinal survey designed for an exclusively Chinese population, to investigate the association between HAP from biomass cooking fuels and the risk of frailty. And we also assessed whether altering cooking fuel types has an effect on frailty risk.

In this study, we found that cooking with biomass fuels was associated with an increased risk of frailty in older adults, as assessed both by the frailty index and the FRAIL scale. And the risk was significantly higher among town residents and participants lacking regular exercise. Switching from biomass to clean fuels might decrease the frailty risk. Our findings add to the existing knowledge of air pollution, and offer evidence on the direct association between HAP from biomass fuel combustion and frailty among older adults.

Exposure to air pollution may cause a series of unfavorable health events (31), and older people are more vulnerable to the effects of air pollution than other age groups (17). Emerging evidence has demonstrated the detrimental effect of ambient air pollution on frailty in older individuals (14–17). However, limited studies have previously explored the potential link between HAP and frailty. A recent cohort, involving 2,225 older adults (mean age 67.9 ± 6.6 years) from the CHARLS, reported that solid cooking fuels are associated with an enhanced risk of phenotypic frailty (adjusted HR 1.26, 95% CI 1.03–1.55) (13), which partly agreed with our present results. Unlike the previous study (13), we enrolled more vulnerable people with older age (mean age 80.9 ± 9.6 years), and the sample size of our cohort was approximately two times that of the previous cohort. In addition to the frail index, we further selected the FRAIL scale, instead of the physical frailty phenotype (PFP) scale, to assess frailty status. Because the FRAIL scale is well established as an excellent, time- and cost-effective screening tool for frailty, which can be rapidly administered in clinical practice without use of performance measurement (as opposed to the PFP scale) (32). Moreover, we conducted a range of stratified and sensitivity analyses to test the validity of our primary results. And an exploratory analysis was also performed to assess the potential impact of switching cooking fuels on frailty risk. Our study revealed that late-life exposure to biomass fuels is a risk factor for frailty, and lowering this exposure might reduce the frailty risk.

Counterintuitively, we demonstrated a stronger association between biomass fuels and frailty among urban residents. Further analysis revealed that the association was only significant in town residents (HR 1.44, 95% CI 1.13–1.84) but not for city dwellers (HR 1.60, 95% CI 0.62–4.15, Supplementary Table S7). One possible explanation may be related to the social structure. In China, towns are considered mixed urban-rural areas (33). Compared to cities, life in towns is relatively simple, traditional, and backward (34). Our present study mirrors this in revealing that about 45% of town residents still use biomass fuels for cooking (Supplementary Table S8). In contrast to those living in rural areas, town residents, however, often have convenient transportation as well as access to more medical services for the acquirement of health diagnoses, which may, in turn, reduce the underestimation of frailty incidence among rural residents. These reasons may help explain the significant association between biomass fuel use and frailty among town residents. Our findings support the potential benefits of accelerating access to clean energy, which is especially promising for public health in low- and middle-income countries that are experiencing an unprecedented pace of urbanization. Our present study also suggests that older adults, who do not perform regular exercise, may exhibit a higher frailty risk when expose to biomass fuels. This may be attributed to the fact that exercise can preserve or improve the function of multiple physiological systems (e.g., neuromuscular and cardiopulmonary function, cognition, and endocrine system) (1), which may partly offset the harmful effect of biomass fuels.

Although statistically non-significant, our exploratory analyses did find that switching cooking fuels from biomass to clean fuels displayed a trend toward reducing the frailty risk. The opposite effect was observed when changing from clean to biomass fuels. These results imply the potential advantages of improving household air environment in preventing or slowing frailty. Contrarily, worsening indoor air environment may exacerbate the frailty conditions among older adults. Given the potential reversibility of frailty (1), our findings may provide the public health implications for creating a healthier aging society in China.

The potential mechanisms underlying biomass fuels and frailty remain unclear, and may be related to fine particulate matter (PM). The burning of biomass fuels produces detrimental chemicals and particles causing HAP, particularly when they are combusted in inefficient cooking stoves (35). It is estimated that the incomplete combustion of solid fuels emits about 10%−38% of their fuel carbon into the air (36). These released substances may induce chronic inflammation, oxidative stress, metabolic alterations, as well as genetic and epigenetic modifications, which can destroy cellular and molecular structures (31), consequently contributing to the cumulative decline in multiple physiological systems. Previous investigations have demonstrated the positive association of HAP from biomass fuel use with cardiovascular diseases (37), respiratory disease (38), cognitive impairment (39), and psychological disorders (40). And the disruption of multi-systemic homeostasis would ultimately contribute to frailty while hampering successful aging.

It is worth noting that household activities such as heating can also cause HAP (38) and result in harmful exposure. Some recent studies have revealed that indoor solid fuels for heating were associated with a higher risk of disability (41) and mortality (42). It can be postulated that the dual exposure to HAP from cooking and heating may exhibit overlapping effects and pose a persistent danger to human health. As data on household heating fuels was lacking in the CLHLS, we were unable to explore the potential impact of heating fuels on frailty. Future studies are needed to take HAP exposure from all domestic activities into consideration. Moreover, mixed use of multiple fuels is common in China, especially in rural families (43). The majority of households prefer to use one primary fuel type while selecting a secondary type as a standby (44). Hence, the primary fuel data may not reflect the true burden of HAP. Co-exposure to chemicals from mixed fuel combustion may cause a synergetic effect (42), and thus mixed fuel use may have more detrimental health impact than a single fuel use alone. And a greater mortality risk of cancer has been observed among mixed wood and coal fuel users, compared with those only using wood (42). Given fuel stacking data was unavailable, the relationship between mixed cooking fuels and frailty was not assessed in this study. The complex interaction between different fuel types requires further research.

With the accelerated economic and societal development, the household fuel use pattern in China may have changed substantially over the past decades. Recent data from the China Family Panel Studies illustrated the considerable change, showing a significant increasing clean cooking fuel and a declining solid cooking fuel usage from 2010 to 2018 (45). This polluting-to-clean transition trend of indoor fuel pattern may bring potential health benefits. However, the data used in this study was about 10 years ago, which may not be used for realistic estimation of household fuel pattern in China today. Hence, the association between biomass cooking fuels and frailty observed in our study might be overrated. Also of note, the lockdown policies and economic downturn associated with the COVID-19 pandemic in recent years may also influence the household energy consumption. The lockdown may increase household usages of solid fuels especially in rural areas (46). Besides, during the lockdown, people were confined to their homes, further extending the exposure time to HAP. It has been reported that lockdown measures during the pandemic significantly increased indoor PM_2.5 exposure (46). Furthermore, COVID-19 infection itself may negatively affect human health. A recent study has indicated that the pandemic decreased the physical activity of older adults and was independently related to a higher incidence of frailty (47). As the CLHLS has not yet released data on 2021 wave, we were unable to explore the potential effect of COVID-19. Our present study may underestimate the true relationship between biomass cooking fuels and frailty following the COVID-19. Further studies are warranted to elucidate the historical change in household fuel pattern and the impact of COVID-19 pandemic.

Our investigation has multiple strengths. First, to our knowledge, this is the largest prospective study examining the association between HAP from biomass fuels and frailty. The large sample size enabled us to carry out expansive subgroup and sensitivity analyses. Second, our study population was older people aged ≥65 years (50.5% of participants ≥80 years). This demographic was vulnerable to air pollution, rarely migrated, and hardly changed their socioeconomic status, thus improving the relevance and generalizability of our results. Third, we applied comprehensive and well-validated assessment tools (both the frailty index and the FRAIL scale) to measure the frailty status, which further strengthens our conclusion. Finally, MI was conducted to deal with missing data, and the complete-cases analyses further supported the reliability of our results. And we also adjusted for multiple risk factors that may affect our results, namely, socioeconomic status, behavior and lifestyle-related factors, geographic region, and other potential confounders, which may aid in the detection of true relationships.

However, some limitations must be acknowledged. First, our participants were exclusively Chinese older adults, limiting the generalizability of our results to young individuals and other populations. Second, household fuel use was self-reported, which may introduce potential recall bias. Besides, we only used primary cooking fuel types as the exposure, and lacking information on fuel stacking may cause exposure misclassification. Directly measure the concentrations of household air pollutants would be more objective. Third, other HAP sources (e.g., fuels use for heating and lighting, and second-hand smoke) and outdoor air pollution were not evaluated due to data unavailability, and therefore residual confounding cannot be excluded. Future studies are required to comprehensively consider the impact of outdoor air pollution and other indoor air pollution sources. Fourth, it was reported that the duration of solid fuels exposure may influence older people's health (48). Nevertheless, in this study, we had insufficient data to precisely assess how long the cooking fuels have been used, and similarly, we also did not know the exact timing of the cooking fuel switch during follow-up, all of which may potentially bias our results. Fifth, the effects of kitchen ventilation facilities and ventilation quality on frailty were not assessed due to lack of relevant data. Six, household fuel choice is strongly related to socioeconomic status (44), which itself can affect the development of frailty (1). Although we have adjusted for several related variables, residual confounding from unmeasured socioeconomic factors may still exist. Finally, the hazardous effect of HAP from biomass fuel usage may be chronic. The relatively short follow-up time of this study (median 2.7 years) may underestimate the association between biomass cooking fuels and frailty. Our study therefore should be considered preliminary. More evidences with longer follow-ups are warranted to examine the possible effects of HAP from biomass fuels, and confirm its association with frailty.

Cooking with biomass fuels was associated with an increased frailty risk in older adults. And this risk was significantly higher among individuals living in town and those lacking regular exercise. Switching from biomass to clean fuels might reduce the frailty risk. Household fuel use pattern in China may have changed dramatically over the past decades with the rapid socioeconomic development and the efforts in policy. The pandemic of COVID-19, instead, may exacerbate household air pollution and pose challenges in access to clean energy, especially in rural areas. Though our present study may not realistically depict the household environment situation in China today, solid fuels (mainly biomass) might currently remain the major fuel for most households in rural China. Our study add evidence on the association between biomass cooking fuels and frailty among older adults. Our findings suggest that improving cooking fuels and access to clean energy might lower the risk of frailty among older people. Further researches are required to validate our results and to elucidate the potential benefits of reducing household biomass fuel use.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Peking University, Beijing, China (IRB00001052-13074). The patients/participants provided their written informed consent to participate in this study.

QSo: conceptualization, methodology, data curation, formal analysis, software, and writing—original draft. MD: visualization, methodology, data curation, software, and writing—review and editing. TL: visualization, methodology, software, and writing—review and editing. YZ: visualization and writing—review and editing. XP and RL: writing—review and editing. QSu: conceptualization, project administration, and writing—review and editing. JY: conceptualization, funding acquisition, project administration, and writing—review and editing. All authors contributed to the article and approved the submitted version.