Next Article in Journal
Adaptation and Validation of the Perceived Restorativeness Scale (PRS) for the Portuguese Population: A Study on the Assessment of the Restorative Effect of Environments
Previous Article in Journal
Proposing a Framework for Ballistic Waste Management in the Context of the Public Security Institute
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Green Health
Volume 1
Issue 2
10.3390/greenhealth1020006
greenhealth-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Habitual Physical Activity and All-Cause Mortality Among Individuals with and Without Impaired Lung Function: Findings from a Prospective Cohort Study

by
Lan Chen
1,
Chongjian Wang
2,
Shiyu Zhang
3,
Shengtao Wei
1,
Jinde Zhao
1 and
Zilong Zhang
1,*
1
Department of Epidemiology, School of Public Health, Sun Yat-sen University, Guangzhou 510080, China
2
Department of Epidemiology and Biostatistics, College of Public Health, Zhengzhou University, Zhengzhou 450001, China
3
Duke Global Health Institute & Nicholas School of the Environment, Duke University, 310 Trent Drive, Durham, NC 27710, USA
*
Author to whom correspondence should be addressed.
Green Health 2025, 1(2), 6; https://doi.org/10.3390/greenhealth1020006
Submission received: 17 March 2025 / Revised: 31 May 2025 / Accepted: 18 June 2025 / Published: 23 June 2025
Browse Figure
Versions Notes

Abstract

Background: The associations between physical activity (PA) and all-cause mortality remain under-investigated among individuals with impaired lung function. Methods: With 201,596 participants from the UK Biobank cohort, baseline pre-bronchodilation lung function tests and a modified International Physical Activity Questionnaire were used to assess lung function status (normal, restricted, obstructed) and PA attributes (volume, intensity, duration). All-cause mortality was determined through linkage to the National Health Services Register. Cox proportional hazard regression was applied to characterize the associations between PA metrics and all-cause mortality among people with different lung function statuses. Dose–response relationships between PA metrics and all-cause mortality risks were examined using restricted cubic splines (number of knots = 4). Results: Over a 11.81-year median follow-up, 5.24% of participants died. All-cause mortality risk declined with increasing total PA volume, plateauing at 1800 MET-min/week without further reduction in individuals with and without impaired lung function. Similar trends were observed for PA intensity and duration, with both factors demonstrating reduced mortality risk that plateaued after reaching a specific threshold. Notably, 24.1% (95% CI: 16.7%, 30.8%) and 43.1% (95% CI: 36.1%, 49.7%) lower mortality risk was observed among individuals with and without impaired lung function for PA with 1201–1800 MET-min/wk. Conclusions: PA was associated with a decreased risk of all-cause mortality among individuals with and without impaired lung function, suggesting that those with impaired lung function might also benefit from PA.

1. Introduction

Habitual physical activity (PA) has been well-documented to be associated with lower premature mortality risk [1,2,3,4]. The World Health Organization recommends that individuals should undertake 150 min of moderate-intensity physical activity, 75 min of vigorous-intensity physical activity, or an equivalent combination of both per week [5]. Several studies even observed further reductions in the risk of developing adverse health outcomes among individuals performing physical activity levels exceeding the recommendations, suggesting engaging in higher physical activity levels may promise additional health benefits [3,6].
However, current evidence concerning the associations between physical activity and mortality risk mainly focused on individuals with normal lung function; very limited studies have investigated the influence of physical activity among people with impaired lung function (restricted or obstructed) [3,7,8]. Impaired lung function has been reported to be an important indicator of premature mortality, as well as a series of chronic diseases [9]. Hence, clarifying the influence of physical activity on mortality among individuals with impaired lung function is vital for guiding effective interventions to alleviate the associated health burden [10].
Nevertheless, studies concerning the associations between physical activity and mortality risk among people with impaired pulmonary function were limited and produced inconsistent results. Supporting evidence from a cohort study with 461,183 participants indicated that about two-thirds of excess mortality risk would be reduced if patients with preserved ratio impaired spirometry performed physical activity meeting the exercise recommendation [11]. Similarly, a 500 MET-min/wk increment in physical activity was reported to be associated with comparable mortality risk reduction among the asthma group (20%) and non-asthma group (18%) [12]. Another systematic review also suggested that patients with chronic obstructive pulmonary disease (COPD) increase physical activity levels in pulmonary rehabilitation to reduce dyspnoea and improve physical function [13]. In an official statement of the European Respiratory Society, it remained obscure whether physical activity improvements may lower mortality risk in COPD patients [14]. Differences in study populations, physical activity measurements, or follow-up duration might be potential reasons for these discrepancies, warranting studies with large sample sizes and standard physical activity measurements to verify the potential associations. Despite the limited and controversial evidence on the influence of physical activity among people with impaired lung function, previous studies mainly investigated the impact of total physical activity volume on death, and few assessed other attributes of physical activity (e.g., intensity and duration), which are also important physical activity components when tailoring recommendations for individuals with different pulmonary function status [4,15].
In the present study, we extracted data from 201,596 participants from the UK Biobank with the aims to (1) characterize the associations between physical activity attributes and all-cause mortality among participants with different lung function statuses and (2) provide evidence to support a physical activity guidance for those with impaired lung function.

2. Methods and Materials

2.1. Study Population

We used data from the UK Biobank study under Application Number 69,550. The UK Biobank study is an ongoing prospective cohort study which recruited over half a million participants aged 38–73 years from 22 assessment centers across England, Wales, and Scotland from 2006 to 2010 [16]. Demographic characteristics, lifestyle factors, biological samples, and health-related outcomes were collected. More details of the study protocols have been documented elsewhere [17,18]. The study was approved by the National Information Governance Board for Health and Social Care and the National Health Service North West Centre for Research Ethics Committee (21/NW/0157). Electronic informed consent was obtained to authorize the use of their de-identified data and linkage to their health-related medical and death records [16]. In the present study, after excluding participants with missing data on lung function (n = 227,696), habitual PA (n = 52,955), used covariates (n = 20,115), and those with outcome records after updated date (n = 99), a total of 201,596 participants were included in the analysis (Supplementary Figure S1).

2.2. Baseline Lung Function Measurement

The Vitalograph Pneumotrac 6800 spirometer (Maids Moreton, UK) was used by trained healthcare technicians and nurses to perform pre-bronchodilation lung function tests. Details of the procedures, including the reproducibility, have been demonstrated elsewhere [19]. Among these reproducible spirograms, the maximum values of forced expiratory volume in one second (FEV1) and forced vital capacity (FVC) from acceptable blows were used to compute the FEV1/FVC ratio [20]. Reference lung volume and the predicted FEV1 were calculated using the Global Lung Initiative (GLI) 2012 equations [21]. Participants were classified into three subgroups according to a modification of the Global Initiative for Chronic Obstructive Lung Disease (GOLD) criteria, including normal (FEV1/FVC ≥ 70%, and FEV1 ≥ 80% predicted), restricted (FEV1/FVC ≥ 70%, and FEV1 < 80% predicted), and obstructed (FEV1/FVC < 70%) [22,23]. Participants were also further classified into two groups (group with normal airflow and group with restricted or obstructed airflow) in the second section of analysis (seen in Section 2.6).

2.3. Physical Activity Measurement

Physical activity (PA) was measured by a modified version of the International Physical Activity Questionnaire (IPAQ) through touchscreen questionnaire at baseline. IPAQ requires participants to report the duration and frequency of walking (low intensity), moderate-intensity, and vigorous-intensity PA undergone in the last 4 weeks. Specific metabolic equivalent (MET, 1 MET = 1 kilocalorie per hour per kilogram of body weight) was assigned to each PA intensity: 3.3 MET for walking, 4.0 MET for moderate-intensity PA, and 8.0 MET for vigorous-intensity PA [24]. The total volume of PA (MET-min/week, abbreviation as MET-min/wk) was then calculated based on the reported intensity, duration, and frequency of PA in one week. Participants were further grouped into three levels according to the criteria of IPAQ [25]: low (<600 MET-min/wk), moderate (600 to 3000 MET-min/wk), and high (≥3000 MET-min/wk). To be noted, 600 MET-min/wk equals performing moderate PA for 150 min per week suggested by the WHO PA guideline [26].
In addition, a total of about 100,000 participants volunteered to wear accelerometers (Axivity AX3, Newcastle upon Tyne, UK) on their dominant wrists to obtain the PA intensity and corresponding duration of each intensity for seven consecutive days [27]. According to the data extraction and processing documented previously [28], a total of 46,070 participants were left and further classified into three groups based on the objectively measured PA tertiles [low (<24.33 milli-gravity (mg)), moderate (24.33 ~ <30.55 mg), and high (≥30.55 mg)]. The procedures for selecting participants with objectively measured PA are listed in Figure S2.

2.4. All-Cause Mortality Ascertainment

Death data were obtained through linkage to the National Health Services (NHS) Register for participants in England and Wales and the NHS Central Register for participants in Scotland [29]. Vital status of all participants was followed till loss of follow-up, death, or 28 February 2021 (the latest date available in the current study), whichever came first [30].

2.5. Covariates

All covariates were selected according to a directed acyclic graph, as well as previous literature. A directed acyclic graph (DAG) [31] developed by a web-based software DAGitty (http://www.dagitty.net/, accessed on 3 May 2023), was applied to help select suitable covariates [32]. We reviewed the previous literature a priori to identify candidate variables that were input into the DAG [33] and then identified the adjustment set of covariates, including demographic characteristics [sex (male and female) and age at recruitment (per 10-year interval)], socioeconomic factors [household income (<18,000 £, 18,000 to 30,999 £, 31,000 to 51,999 £, 52,000 to 100,000 £, >100,000 £, and unknown), education (higher degree, any school degree, vocational qualifications, and other), employment status (paid, retired, and unpaid), and Townsend deprivation index (TDI, quintile-classified)] and lifestyle factors [diet (unhealthy, moderate, healthy, and unknown), smoking status (never, previous, and current smokers), alcohol intake frequency (never, occasional, moderate, and heavy), sedentary time (low, moderate, high, and unknown), body mass index (BMI, underweight, normal, overweight, and obese), and secondhand smoke exposure (no, yes, and unknown)]. A simplified version of DAG is displayed in Figure S3. And details of the definitions of these covariates are shown in Table S1.

2.6. Statistical Analysis

Cox proportional hazard regression models with follow-up time as time scale were used to estimate the hazard ratios and their corresponding 95% confidence intervals (HRs, 95% CIs) for the associations of PA attributes and lung function with all-cause mortality. Schoenfeld residuals were plotted to examine the proportional hazard assumption, and no evidence of violation was found. Estimates were calculated per 600 MET-min/wk increment in PA volume when PA volume was analyzed as a continuous variable. In addition, low PA level and normal lung function were presented as referent groups when these two factors were treated as categorical variables. Considering the potential changes during follow-up, changes in PA volume between the baseline (2006–2010) and the second resurvey (2014) were analyzed (Supplementary Methods) [34].
In the first section, independent influences of PA volume and lung function on all-cause death were examined. Three incremental models were fitted, including Model 1, adjusted for demographic characteristics; Model 2, which further adjusted for socioeconomic and lifestyle factors; and Model 3, which included PA volume and lung function simultaneously for mutual adjustment. Dose–response relationships between three PA metrics and all-cause mortality risks were examined using restricted cubic splines (number of knots = 4) to assess the potential non-linearity. According to a PA–mortality plateau curve observed in previous similar research, we further conducted the segmented Cox regressions with the cut-off value of potential plateau threshold to investigate the Pfor trend value before and after the threshold [1]. Pfor trend less than 0.025 (Bonferroni adjusted) was considered to have decreasing trend. We further classified participants into nine groups based on PA volume level (low, moderate, high) and lung function categories (normal, restricted, obstructed). Low PA level and normal lung function were treated as reference group. Possible interactions between PA volume and lung function were also assessed (shown in Supplementary Methods).
In the second section, we further investigated the associations between the intensity and duration of PA and all-cause mortality among participants with and without impaired lung function. First, unlike the relatively rough cut-off value defined by current PA guidelines [5], we additionally divided the total PA volume into 10 groups to find suitable PA volume against mortality risk. Low-to-moderate intensity of PA (LMPA) and vigorous intensity of PA (VPA) were identified according to self-reported PA intensity. The duration of LMPA was calculated and classified into 8 groups. Similarly, the duration of VPA was classified into 6 groups. Because only 3329 (1.7%) participants reported not taking PA over the entire follow-up, the referent group for PA volume was broadened to 0–330 MET-min/wk (about 10% of the distribution) with reference to a previous, similar study [3]. The proportion of VPA volume to total PA volume was generated to examine whether participants would experience lower mortality risk if performing higher proportion of VPA [35]. Details of the classification of LMPA duration, VPA duration, and VPA proportion are listed in Table S2.

2.7. Sensitivity Analysis

We conducted several sensitivity analyses to test the robustness of our results. First, to control the potential influence of baseline health condition [36], we additionally adjusted for the baseline multimorbidity (a binary variable shown in Supplementary Methods). Participants would be defined to have multimorbidity if they have two or more diseases at the same time [37]. Second, we excluded those who died within the first 2 years since enrollment to avoid potential reverse causality. Third, instead of excluding participants with missing covariates, we applied multiple imputation via chained equation (MICE) (the number of imputations was 5) [38]. Fourth, to enhance clinical implications of our findings, we extracted data from the first-occurrence record in the UK Biobank [39] and ascertained a total of 10,295 asthma and 1799 COPD cases among those with impaired lung function at baseline. We investigated the associations between three PA attributes (total PA volume, duration, intensity) and mortality risk across participants with asthma or COPD separately.
Statistical analyses were performed by R.4.0.1 (Vienna, Austria). A two-sided p value of less than 0.05 was considered statistically significant.

3. Results

3.1. Baseline Characteristics of Study Population

Participants’ characteristics are presented in Table 1. Of the 201,596 participants, 10,560 (5.24%) deceased cases were observed during the follow-up (mean follow-up of 11.81 years). The mean baseline age of all participants was 56.38 [standard deviation (SD): 7.99] years, and there were more females (56.18%) than males. The proportion of subjects with normal, restricted, and obstructed airflow limitation was 80.16%, 6.74%, and 13.10%, respectively. Participants with restricted airflow were less likely than those with normal airflow to have healthy BMI, healthy diet, or moderate-to-high intensity PA.

3.2. Associations of PA and Lung Function with All-Cause Mortality

Results of the independent associations of PA and lung function with all-cause death are displayed in Table 2. Generally, increased PA level was associated with decreasing mortality risk, and the associations remained statistically significant in Model 3 (fully adjusted model). The HRs (95% CIs) were 0.824 (0.782, 0.867) and 0.767 (0.727, 0.808) for participants with moderate and high PA levels relative to those with low PA levels. In contrast, declined airflow was associated with higher mortality risk, with HRs (95% CIs) of 1.439 (1.348, 1.535) and 1.346 (1.282, 1.412) for individuals with restricted and obstructed airflow in fully adjusted models, respectively.
The inverse relationships between PA and all-cause mortality remained stable regardless of lung function status, and the associations were even stronger among participants with restricted or obstructed airflow. For example, among participants with restricted airflow, the HRs (95% CIs) were 0.727 (0.629, 0.840) and 0.650 (0.556, 0.761) for moderate and high PA levels relative to low PA levels, which were smaller than corresponding HRs observed in individuals with normal lung function. In line with self-reported PA, increasing the objectively measured PA level was also associated with decreasing mortality risk, and individuals with restricted or obstructed lung function still experienced relatively lower mortality risks relative to those with normal lung function given the same PA level (Table S3). Statistically significant antagonism between decreasing lung function and increasing PA level was found on both additive and multiplicative scales, with nearly all RERI values < 0 and the p value of multiplicative interaction term < 0.05 (Table 3).

3.3. Associations Between PA Attributes and All-Cause Mortality

The dose–response relationship between total PA volume and all-cause mortality is shown in Figure 1. Participants with and without impaired lung function shared similar dose–response curves, as the mortality risk decreased up to 1800 MET-min/wk and then reached a plateau. Specifically, for participants with normal lung function, the HRs decreased to 0.759 and, subsequently, fluctuated around 0.770 once PA volume exceeded 1800 MET-min/wk. Similarly, for those with impaired lung function, HRs decreased to 0.569 and stabilized around 0.580 once PA exceeded 1800 MET-min/week. The plateau in the PA–mortality risk curve was also supported by Pfor trend values: for PA volumes ≤ 1800 MET-min/week, Pfor trend = 0.036 (normal lung function) and Pfor trend = 0.001 (impaired lung function); and for PA volumes > 1800 MET-min/week, Pfor trend = 0.970 (normal lung function) and Pfor trend = 0.280 (impaired lung function). Notably, participants with impaired lung function (HR: 0.569, 95% CI: 0.503, 0.643) experienced a larger reduction in mortality risk than those with normal lung function (HR: 0.759, 95% CI: 0.692, 0.833). Similar results were observed when PA was further grouped by a finer scale, as the mortality risks decreased between 0 and 1800 MET-min/wk and then reached a relatively stable status (Table S4).
In terms of other attributes of PA (intensity and duration), a similar plateau of mortality risk reduction was observed. For LMPA, the mortality risk progressively decreased with increasing duration across all lung function groups, plateauing with minor fluctuations beyond 450 min/week. Regarding VPA duration, individuals with restricted or obstructed airflow exhibited the lowest mortality risk at 10–74 min/wk. Longer VPA durations (≥75 min/wk) showed no association with further risk reduction. Among those with normal airflow, the lowest mortality risk was observed for taking PA up to 150 min/wk. Consistently, the VPA proportion was also associated with a decreased mortality risk, as the proportion of less than 25% showed the lowest mortality risk, and no further reduction was observed for a proportion over 25%. Lower mortality risks were observed among people with restricted or obstructed airflow relative to those with normal airflow, with HRs (95% CIs) of 0.859 (0.81, 0.911), 0.668 (0.465, 0.960), and 0.705 (0.633, 0.786) for individuals with normal, restricted, and obstructed airflow with less than 25% VPA proportion, respectively (Table 4). Dose–response relationships between PA attributes (LMPA duration, VPA duration, and VPA proportion) and mortality risk are shown in Supplementary Figures S4–S6. Generally, the mortality risk decreased with increasing duration of LMPA, though a slight risk increase emerged at 450 min/wk. In contrast, VPA showed a U-shaped dose–response relationship with mortality, with the lowest risk observed at 150 min/wk.

3.4. Sensitivity Analyses

All sensitivity analyses including those adjusted for baseline health conditions, excluding participants dying in the first two years since recruitment, and conducted with multiple imputation of missing covariates yielded similar results to our main findings (Tables S4 and S5). In addition, the inverse association between physical activity (PA) attributes and mortality risk was consistent across individuals with asthma and COPD, where higher PA levels correlated with reduced mortality. The lowest mortality risk occurred with 600–900 min/week LMPA, 75–150 min/week VPA, or ≥25% VPA of total PA (Table S6). Subgroup analysis by smoking status indicated that the mortality risk decreased along with increasing PA volume, and the estimates were stronger among COPD patients who never smoked. In asthma cases, the associations between PA volume and mortality risk shared the same pattern, and the estimates did not differ in smokers and non-smokers (Table S7).

4. Discussion

In this large prospective cohort study, we found that better lung function or higher PA level was associated with lower mortality risk. The inverse relationships between PA and all-cause mortality remained stable across three lung function categories. Notably, a significant antagonism between increasing PA level and decreasing lung function was observed, indicating taking PA actively might mitigate the influence of impaired lung function on mortality risk. Adults with impaired lung function might obtain relative benefits to those with normal lung function for a given PA level. Our results further suggested that PA–mortality associations followed a plateau pattern, with no significant risk reduction beyond 1800 MET-min/week.
Consistent with our findings, a large body of studies have reported the negative associations between PA and all-cause mortality [40,41], as well as the positive relationships between declined lung function and death [42,43]. Our study confirmed existing evidence of the health benefits of PA and the adverse influence of impaired lung function on death. Similarly to previous findings, we observed a significant antagonism between declined lung function and increasing PA level, with the values of relative excess risk due to interaction (RERI) less than 0. A negative RERI value in the present study implied that increasing PA level and declining lung function act antagonistically on all-cause mortality, and increasing PA level alleviated the adverse influence of impaired lung function on mortality [44]. Consistently, Shu et al. also indicated that PA could mitigate the adverse impacts of COPD on death, and the risk of mortality was comparable between fully active COPD patients and low-active control subjects [45]. Wouters et al. also found that increasing PA levels reduced mortality risk in patients with obstructive airflow [46]. Several cross-sectional studies demonstrated the mitigating role of PA in declining lung function as well [47,48]. All these findings implied that PA still played a beneficial role in lowering mortality risk for those with declined pulmonary function. Concerning the limitations of observational studies and the heterogeneity in measurements of PA and lung function, future biological studies concerning the interplay of PA and lung function are warranted to investigate the underlying mechanisms of the antagonism of these two factors.
Of note, we observed that participants with impaired lung function might achieve comparable mortality risk reduction to those with normal lung function at equivalent PA levels. Similarly, another cohort study recruiting about 18,000 participants reported that per 500 MET-min/wk increment in PA volume was associated with a 20% and 18% reduction in all-cause mortality among participants with and without asthma [12]. Evidence from the Korean National Health Insurance Service-National Sample Cohort indicated that COPD patients engaging in regular physical activity exhibited mortality risks comparable to or lower than inactive non-COPD individuals [45]. Furthermore, analysis of the MJ cohort (n ≈ 40,000) demonstrated that adults with preserved ratio impaired spirometry who met PA guidelines experienced a 66% reduction in mortality risk (absolute risk decrease: 28% to 9%) [11]. Direct comparison of our findings with existing evidence was limited by insufficient prior research on heterogeneity/homogeneity patterns. Nevertheless, the current literature on PA–lung function associations provides partial context for our results. Impaired lung function does not inherently predict reduced physical activity [36]. Regular PA engagement may attenuate risks of accelerated lung function decline and premature mortality [49], supporting recommendations for exercise training to enhance pulmonary rehabilitation and physical fitness in this population [50]. However, our findings were derived from observational data, and residual confounding remains possible. Further interventional studies are warranted to verify our results.
Notably, a plateau effect characterized the dose–response relationships between mortality risk reduction and increasing total physical activity (PA) volume, as well as other PA attributes (intensity, duration). Specifically, mortality risk decreased with PA volumes up to 1800 MET-min/week, plateauing beyond this threshold (Figure 1, Table S4)—aligning with the prior literature [1,3,51]. For instance, a meta-analysis with 661,137 participants observed benefit saturation at 1350–2400 MET-min/week of leisure-time PA, with no additional survival advantage at higher volumes [51]. Similarly, a narrative review also indicated that maximal risk reduction at about 3000 MET-min/wk (substantially exceeding current PA recommendations), beyond which no further mortality benefits accrued [52]. This plateau of morality risk reduction may reflect diminishing physiological returns in older adults at higher PA intensities. It was reported that the maximum oxygen consumption would be improved and then stabilized with increasing PA volume, suggesting that cardiorespiratory fitness gains, a key mediator of mortality reduction, reach a ceiling beyond certain activity thresholds [53].
Regarding the duration of PA with specific intensities, a similar plateau was found as a longer duration of PA might not promise a further reduction in mortality risk. We proposed thresholds of 450 min/wk for LMPA or 150 min/wk for VPA as potential targets for those with impaired lung function. This aligns with a prospective cohort of 116,221 American adults who reported no more survival benefits beyond 450 min/wk of moderate–intensity PA [3]. Importantly, given the heterogeneity in study populations, methods used to measure physical activity and lung function, as well as unmeasured confounding, our findings should be interpreted and extrapolated with caution. Further studies using objectively measured PA are needed to verify our results.
Though the exact mechanisms concerning the lower mortality risk for engaging in PA still remain not fully understood, improvements in cardiorespiratory fitness due to PA could be a possible mechanism. Moderate–to–vigorous PA could increase peak oxygen consumption to improve cardiorespiratory fitness [54]. In addition, PA could also regulate lipid and glucose metabolism, immune function, and lower inflammation levels, which were critical in maintaining cardiorespiratory health [55]. Better cardiorespiratory fitness, in return, promised the ability to engage in PA actively.

Strengths and Limitations

Our study possesses several strengths. First, our study is among the first to investigate whether individuals with impaired lung function achieve comparable mortality reduction to those with normal lung function at equivalent physical activity levels across multiple PA attributes. These insights may guide personalized activity prescriptions based on respiratory status. Second, leveraging the UK Biobank cohort with extensive covariates adjustment enabled robust characterization of associations between lung function, PA characteristics, and all-cause mortality. Several limitations also need to be noted. First, as data on PA and covariates were collected at baseline, potential changes during follow-up might have influenced the estimates. However, we further carried out an additional analysis among participants (n = 19,454) who attended both the baseline survey and 2014 resurvey and found that about 63% of participants remained at their PA level, and the distribution of PA levels between people with and without lung function impairment was comparable (Table S8), implying that the changes in PA were non-differentially distributed, and this change would not influence our results substantially. Second, all PA information was collected from self-reported questionnaires, recall bias could not be avoided, which could influence our results inevitably. We further investigated the associations between objectively measured PA and mortality risk and observed consistent results for self-reported PA (Table S3). Third, almost 50% of the total population was excluded due to missing data, which implied that the findings should be interpreted with caution as the representativeness was inevitably influenced. Moreover, we conducted multivariate imputation via chained equation (MICE), and no substantial variation in results was observed (Tables S4 and S5). In addition, we compared demographic characteristics between included and excluded participants and observed that the distribution of characteristics (e.g., sex and age) was comparable among the two groups (Table S9). Fourth, healthy volunteer bias has been widely reported in the UK Biobank [56]. Relatively healthier volunteers in the UK Biobank than the general population may overestimate PA benefits and underrepresent severe lung impairment. In addition, this might affect the generalizability of our results to more diverse or sicker populations, such as those in low- and middle-income countries where lung function impairment may be more prevalent due to environmental factors. Future research from different regions is needed to verify our findings. Fifth, post-bronchodilation spirometry tests were not conducted in the UK Biobank, which precluded an accurate diagnosis of asthma or COPD. Using pre-bronchodilation spirometry data only might have misclassified obstructive diseases (e.g., asthma or COPD) to some extent [57].

5. Conclusions

Our analysis revealed that both impaired lung function and lower PA levels independently predicted higher mortality risk across all lung function statuses. Mortality risk decreased progressively with higher PA volumes, plateauing beyond 1800 MET-min/week. This suggests that 450 min/wk of moderate-intensity PA may represent an optimal threshold for individuals with impaired lung function. Intervention studies in real-world populations are warranted to validate these findings.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/greenhealth1020006/s1, Supplementary methods: Table S1: Definitions of covariates used in the analysis; Table S2: Information of classifications of total PA volume, LMPA duration, VPA duration and VPA proportion; Supplementary Results: Table S3: Associations between objectively-measured physical activity and mortality among people with different lung function categories; Table S4: Associations between total physical activity volume (MET–min/wk) and all-cause mortality among subjects with or without impaired lung function; Table S5: Sensitivity analysis of the associations between duration of LMPA, VPA, and all-cause mortality among subjects with or without impaired lung function; Table S6: Associations between PA attributes and mortality risk among participants diagnosed with asthma or COPD at baseline; Table S7: Associations between PA volume and mortality risk across different smoking status in asthma and COPD cases separately; Table S8: Changes in PA levels between 2006 and 2010 baseline and resurvey between 2014 and 2020 by lung function status among 19,454 participants; Table S9: Comparison of baseline characteristics between included and excluded participants; Figure S1: Flowchart of participants selection; Figure S2: Selection of participants with objectively measured physical activity; Figure S3: Directed acyclic graph (DAG) model used for covariate selection; Figure S4: Dose–response relationship between duration of LMPA and mortality risk across three lung function categories; Figure S5: Dose–response relationship between duration of VPA and mortality risk across three lung function categories; Figure S6: Dose–response relationship between proportion of VPA volume to total PA volume and mortality risk across three lung function categories. References [58,59,60,61,62,63] are cited in the supplementary materials.

Author Contributions

L.C. conceived the idea and contributed to the analysis, interpretation, and drafting of the work. C.W. and S.Z. contributed to the validation of results. S.W. and J.Z. contributed to the draft review. Z.Z. contributed to the acquisition of data, supervision of the work, and editing of the manuscript. All authors have read and agreed to the published version of the manuscript.

Funding

None applicable.

Institutional Review Board Statement

The study was approved by the National Information Governance Board for Health and Social Care and the National Health Service North West Centre for Research Ethics Committee (21/NW/0157).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

Data from the UK Biobank (https://www.ukbiobank.ac.uk/, accessed on 14 January 2021), the Health Surveys for England, and the Scottish Health Surveys (https://data-archive.ac.uk/, accessed on 14 January 2021) are available to bona fide researchers upon application. Detailed data application process is listed in UK Biobank webpage (https://www.ukbiobank.ac.uk/enable-your-research/apply-for-access, accessed on 14 January 2021).

Acknowledgments

We wish to acknowledge the UK Biobank Resource under Application Number 69550 and also gratefully acknowledge the participants and the valuable contributions of members of the UK Biobank center.

Conflicts of Interest

All authors declared no competing interests.

References

  1. Blond, K.; Brinklov, C.F.; Ried-Larsen, M.; Crippa, A.; Grontved, A. Association of high amounts of physical activity with mortality risk: A systematic review and meta-analysis. Br. J. Sports Med. 2020, 54, 1195–1201. [Google Scholar] [CrossRef] [PubMed]
  2. Ding, D.; Van Buskirk, J.; Nguyen, B.; Stamatakis, E.; Elbarbary, M.; Veronese, N.; Clare, P.J.; Lee, I.M.; Ekelund, U.; Fontana, L. Physical activity, diet quality and all-cause cardiovascular disease and cancer mortality: A prospective study of 346 627 UK Biobank participants. Br. J. Sports Med. 2022, 56, 1148–1156. [Google Scholar] [CrossRef]
  3. Lee, D.H.; Rezende, L.F.M.; Joh, H.K.; Keum, N.; Ferrari, G.; Rey-Lopez, J.P.; Rimm, E.B.; Tabung, F.K.; Giovannucci, E.L. Long-Term Leisure-Time Physical Activity Intensity and All-Cause and Cause-Specific Mortality: A Prospective Cohort of US Adults. Circulation 2022, 146, 523–534. [Google Scholar] [CrossRef] [PubMed]
  4. Burtscher, J.; Millet, G.P.; Gatterer, H.; Vonbank, K.; Burtscher, M. Does Regular Physical Activity Mitigate the Age-Associated Decline in Pulmonary Function? Sports Med. 2022, 52, 963–970. [Google Scholar] [CrossRef]
  5. Bull, F.C.; Al-Ansari, S.S.; Biddle, S.; Borodulin, K.; Buman, M.P.; Cardon, G.; Carty, C.; Chaput, J.P.; Chastin, S.; Chou, R.; et al. World Health Organization 2020 guidelines on physical activity and sedentary behaviour. Br. J. Sports Med. 2020, 54, 1451–1462. [Google Scholar] [CrossRef] [PubMed]
  6. Vitti, A.; Nikolaidis, P.T.; Villiger, E.; Onywera, V.; Knechtle, B. The “New York City Marathon”: Participation and performance trends of 1.2M runners during half-century. Res. Sports Med. 2020, 28, 121–137. [Google Scholar] [CrossRef]
  7. Shiroma, E.J.; Sesso, H.D.; Moorthy, M.V.; Buring, J.E.; Lee, I.M. Do moderate-intensity and vigorous-intensity physical activities reduce mortality rates to the same extent? J. Am. Heart Assoc. 2014, 3, e000802. [Google Scholar] [CrossRef] [PubMed]
  8. Sabia, S.; Dugravot, A.; Kivimaki, M.; Brunner, E.; Shipley, M.J.; Singh-Manoux, A. Effect of intensity and type of physical activity on mortality: Results from the Whitehall II cohort study. Am. J. Public Health 2012, 102, 698–704. [Google Scholar] [CrossRef]
  9. Cho, S.J.; Stout-Delgado, H.W. Aging and Lung Disease. Annu. Rev. Physiol. 2020, 82, 433–459. [Google Scholar] [CrossRef]
  10. Divo, M.J.; Marin, J.M.; Casanova, C.; Cabrera Lopez, C.; Pinto-Plata, V.M.; Marin-Oto, M.; Polverino, F.; de-Torres, J.P.; Billheimer, D.; Celli, B.R.; et al. Comorbidities and mortality risk in adults younger than 50 years of age with chronic obstructive pulmonary disease. Respir. Res. 2022, 23, 267. [Google Scholar] [CrossRef]
  11. Wijnant, S.R.A.; De Roos, E.; Kavousi, M.; Stricker, B.H.; Terzikhan, N.; Lahousse, L.; Brusselle, G.G. Trajectory and mortality of preserved ratio impaired spirometry: The Rotterdam Study. Eur. Respir. J. 2020, 55, 1901217. [Google Scholar] [CrossRef] [PubMed]
  12. Lee, H.; Ryu, J.; Chung, S.J.; Park, D.W.; Park, T.S.; Moon, J.Y.; Kim, T.H.; Sohn, J.W.; Yoon, H.J.; Kim, S.H. Overall and respiratory mortality reduction with physical activity in subjects with and without asthma. Allergy 2023, 78, 1677–1680. [Google Scholar] [CrossRef]
  13. Østergaard, E.B.; Sritharan, S.S.; Kristiansen, A.D.; Thomsen, P.M.; Løkke, A. Barriers and motivational factors towards physical activity in daily life living with COPD—An interview based pilot study. Eur. Clin. Respir. J. 2018, 5, 1484654. [Google Scholar] [CrossRef] [PubMed]
  14. Watz, H.; Pitta, F.; Rochester, C.L.; Garcia-Aymerich, J.; ZuWallack, R.; Troosters, T.; Vaes, A.W.; Puhan, M.A.; Jehn, M.; Polkey, M.I.; et al. An official European Respiratory Society statement on physical activity in COPD. Eur. Respir. J. 2014, 44, 1521–1537. [Google Scholar] [CrossRef]
  15. Wen, C.P.; Wai, J.P.; Tsai, M.K.; Yang, Y.C.; Cheng, T.Y.; Lee, M.C.; Chan, H.T.; Tsao, C.K.; Tsai, S.P.; Wu, X. Minimum amount of physical activity for reduced mortality and extended life expectancy: A prospective cohort study. Lancet 2011, 378, 1244–1253. [Google Scholar] [CrossRef]
  16. Elovainio, M.; Hakulinen, C.; Pulkki-Råback, L.; Virtanen, M.; Josefsson, K.; Jokela, M.; Vahtera, J.; Kivimäki, M. Contribution of risk factors to excess mortality in isolated and lonely individuals: An analysis of data from the UK Biobank cohort study. Lancet Public Health 2017, 2, e260–e266. [Google Scholar] [CrossRef]
  17. Celis-Morales, C.A.; Welsh, P.; Lyall, D.M.; Steell, L.; Petermann, F.; Anderson, J.; Iliodromiti, S.; Sillars, A.; Graham, N.; Mackay, D.F.; et al. Associations of grip strength with cardiovascular, respiratory, and cancer outcomes and all cause mortality: Prospective cohort study of half a million UK Biobank participants. BMJ 2018, 361, k1651. [Google Scholar] [CrossRef] [PubMed]
  18. Williams, D.M.; Jylhävä, J.; Pedersen, N.L.; Hägg, S. A Frailty Index for UK Biobank Participants. J. Gerontol. Ser. A. Biol. Sci. Med. Sci. 2019, 74, 582–587. [Google Scholar] [CrossRef]
  19. Miller, M.R.; Hankinson, J.; Brusasco, V.; Burgos, F.; Casaburi, R.; Coates, A.; Crapo, R.; Enright, P.; van der Grinten, C.P.; Gustafsson, P.; et al. Standardisation of spirometry. Eur. Respir. J. 2005, 26, 319–338. [Google Scholar] [CrossRef]
  20. Doiron, D.; de Hoogh, K.; Probst-Hensch, N.; Fortier, I.; Cai, Y.; De Matteis, S.; Hansell, A.L. Air pollution, lung function and COPD: Results from the population-based UK Biobank study. Eur. Respir. J. 2019, 54, 1802140. [Google Scholar] [CrossRef]
  21. Quanjer, P.H.; Stanojevic, S.; Cole, T.J.; Baur, X.; Hall, G.L.; Culver, B.H.; Enright, P.L.; Hankinson, J.L.; Ip, M.S.; Zheng, J.; et al. Multi-ethnic reference values for spirometry for the 3-95-yr age range: The global lung function 2012 equations. Eur. Respir. J. 2012, 40, 1324–1343. [Google Scholar] [CrossRef] [PubMed]
  22. Rabe, K.F.; Hurd, S.; Anzueto, A.; Barnes, P.J.; Buist, S.A.; Calverley, P.; Fukuchi, Y.; Jenkins, C.; Rodriguez-Roisin, R.; van Weel, C.; et al. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am. J. Respir. Crit. Care Med. 2007, 176, 532–555. [Google Scholar] [CrossRef] [PubMed]
  23. Mannino, D.M.; Diaz-Guzman, E. Interpreting lung function data using 80% predicted and fixed thresholds identifies patients at increased risk of mortality. Chest 2012, 141, 73–80. [Google Scholar] [CrossRef]
  24. Craig, C.L.; Marshall, A.L.; Sjöström, M.; Bauman, A.E.; Booth, M.L.; Ainsworth, B.E.; Pratt, M.; Ekelund, U.; Yngve, A.; Sallis, J.F.; et al. International physical activity questionnaire: 12-country reliability and validity. Med. Sci. Sports Exerc. 2003, 35, 1381–1395. [Google Scholar] [CrossRef]
  25. IPAQ Research Committee. Guidelines for Data Processing and Analysis of the International Physical Activity Questionnaire (IPAQ)—Short and Long Forms. 2005. Available online: https://biobank.ndph.ox.ac.uk/showcase/ukb/docs/ipaq_analysis.pdf (accessed on 16 November 2022).
  26. WHO. WHO Guidelines on Physical Activity and Sedentary Behaviour. 2020. Available online: https://iris.who.int/bitstream/handle/10665/336656/9789240015128-eng.pdf?sequence=1 (accessed on 13 April 2023).
  27. Hamer, M.; Sharma, N.; Batty, G.D. Association of objectively measured physical activity with brain structure: UK Biobank study. J. Intern. Med. 2018, 284, 439–443. [Google Scholar] [CrossRef]
  28. Ramakrishnan, R.; Doherty, A.; Smith-Byrne, K.; Rahimi, K.; Bennett, D.; Woodward, M.; Walmsley, R.; Dwyer, T. Accelerometer measured physical activity and the incidence of cardiovascular disease: Evidence from the UK Biobank cohort study. PLoS Med. 2021, 18, e1003487. [Google Scholar] [CrossRef]
  29. Chudasama, Y.V.; Zaccardi, F.; Gillies, C.L.; Dhalwani, N.N.; Yates, T.; Rowlands, A.V.; Davies, M.J.; Khunti, K. Leisure-time physical activity and life expectancy in people with cardiometabolic multimorbidity and depression. J. Intern. Med. 2020, 287, 87–99. [Google Scholar] [CrossRef]
  30. Emerging Risk Factors Collaboration; Angelantonio, E.D.; Kaptoge, S.; Wormser, D.; Willeit, P.; Butterworth, A.S.; Bansal, N.; O’Keeffe, L.M.; Gao, P.; Wood, A.M.; et al. Association of Cardiometabolic Multimorbidity With Mortality. JAMA 2015, 314, 52–60. [Google Scholar] [CrossRef]
  31. Tennant, P.W.G.; Murray, E.J.; Arnold, K.F.; Berrie, L.; Fox, M.P.; Gadd, S.C.; Harrison, W.J.; Keeble, C.; Ranker, L.R.; Textor, J.; et al. Use of directed acyclic graphs (DAGs) to identify confounders in applied health research: Review and recommendations. Int. J. Epidemiol. 2021, 50, 620–632. [Google Scholar] [CrossRef]
  32. Ramirez, F.D.; Chen, S.; Langan, S.M.; Prather, A.A.; McCulloch, C.E.; Kidd, S.A.; Cabana, M.D.; Chren, M.M.; Abuabara, K. Association of Atopic Dermatitis With Sleep Quality in Children. JAMA Pediatr. 2019, 173, e190025. [Google Scholar] [CrossRef]
  33. Evandt, J.; Oftedal, B.; Krog, N.H.; Skurtveit, S.; Nafstad, P.; Schwarze, P.E.; Skovlund, E.; Houthuijs, D.; Aasvang, G.M. Road traffic noise and registry based use of sleep medication. Environ. Health 2017, 16, 110. [Google Scholar] [CrossRef] [PubMed]
  34. Han, Y.; Hu, Y.; Yu, C.; Guo, Y.; Pei, P.; Yang, L.; Chen, Y.; Du, H.; Sun, D.; Pang, Y.; et al. Lifestyle, cardiometabolic disease, and multimorbidity in a prospective Chinese study. Eur. Heart J. 2021, 42, 3374–3384. [Google Scholar] [CrossRef]
  35. Wang, Y.; Nie, J.; Ferrari, G.; Rey-Lopez, J.P.; Rezende, L.F.M. Association of Physical Activity Intensity With Mortality: A National Cohort Study of 403 681 US Adults. JAMA Intern. Med. 2021, 181, 203–211. [Google Scholar] [CrossRef] [PubMed]
  36. Clarenbach, C.F.; Sievi, N.A.; Haile, S.R.; Brack, T.; Brutsche, M.H.; Frey, M.; Irani, S.; Leuppi, J.D.; Thurnheer, R.; Kohler, M. Determinants of annual change in physical activity in COPD. Respirology 2017, 22, 1133–1139. [Google Scholar] [CrossRef]
  37. Arokiasamy, P.; Uttamacharya, U.; Jain, K.; Biritwum, R.B.; Yawson, A.E.; Wu, F.; Guo, Y.; Maximova, T.; Espinoza, B.M.; Rodríguez, A.S.; et al. The impact of multimorbidity on adult physical and mental health in low- and middle-income countries: What does the study on global ageing and adult health (SAGE) reveal? BMC Med. 2015, 13, 178. [Google Scholar] [CrossRef]
  38. Ho, F.K.; Gray, S.R.; Welsh, P.; Petermann-Rocha, F.; Foster, H.; Waddell, H.; Anderson, J.; Lyall, D.; Sattar, N.; Gill, J.M.R.; et al. Associations of fat and carbohydrate intake with cardiovascular disease and mortality: Prospective cohort study of UK Biobank participants. BMJ 2020, 368, m688. [Google Scholar] [CrossRef] [PubMed]
  39. Li, S.; Xiang, Y.; Yang, X.; Chen, J.; Xian, W.; Wang, Y. Associations of sugary beverage consumption with chronic obstructive pulmonary disease, asthma, and asthma-chronic obstructive pulmonary disease overlap syndrome: A prospective cohort study. Am. J. Clin. Nutr. 2024, 120, 707–718. [Google Scholar] [CrossRef]
  40. Watts, E.L.; Matthews, C.E.; Freeman, J.R.; Gorzelitz, J.S.; Hong, H.G.; Liao, L.M.; McClain, K.M.; Saint-Maurice, P.F.; Shiroma, E.J.; Moore, S.C. Association of Leisure Time Physical Activity Types and Risks of All-Cause, Cardiovascular, and Cancer Mortality Among Older Adults. JAMA Netw. Open 2022, 5, e2228510. [Google Scholar] [CrossRef]
  41. Zhao, M.; Veeranki, S.P.; Li, S.; Steffen, L.M.; Xi, B. Beneficial associations of low and large doses of leisure time physical activity with all-cause, cardiovascular disease and cancer mortality: A national cohort study of 88,140 US adults. Br. J. Sports Med. 2019, 53, 1405–1411. [Google Scholar] [CrossRef]
  42. Baughman, P.; Marott, J.L.; Lange, P.; Martin, C.J.; Shankar, A.; Petsonk, E.L.; Hnizdo, E. Combined effect of lung function level and decline increases morbidity and mortality risks. Eur. J. Epidemiol. 2012, 27, 933–943. [Google Scholar] [CrossRef]
  43. Garcia-Aymerich, J.; Serra Pons, I.; Mannino, D.M.; Maas, A.K.; Miller, D.P.; Davis, K.J. Lung function impairment, COPD hospitalisations and subsequent mortality. Thorax 2011, 66, 585–590. [Google Scholar] [CrossRef] [PubMed]
  44. Chen, L.; Cai, M.; Li, H.; Wang, X.; Tian, F.; Wu, Y.; Zhang, Z.; Lin, H. Risk/benefit tradeoff of habitual physical activity and air pollution on chronic pulmonary obstructive disease: Findings from a large prospective cohort study. BMC Med. 2022, 20, 70. [Google Scholar] [CrossRef]
  45. Shu, C.C.; Lee, J.H.; Tsai, M.K.; Su, T.C.; Wen, C.P. The ability of physical activity in reducing mortality risks and cardiovascular loading and in extending life expectancy in patients with COPD. Sci. Rep. 2021, 11, 21674. [Google Scholar] [CrossRef] [PubMed]
  46. Wouters, E.F.M.; Franssen, F.M.E.; Spruit, M.A. Survival and physical activity in COPD: A giant leap forward! Chest 2011, 140, 279–281. [Google Scholar] [CrossRef] [PubMed]
  47. Chen, H.I.; Kuo, C.S. Relationship between respiratory muscle function and age, sex, and other factors. J. Appl. Physiol. 1989, 66, 943–948. [Google Scholar] [CrossRef]
  48. Benadjaoud, M.A.; Menai, M.; van Hees, V.T.; Zipunnikov, V.; Regnaux, J.P.; Kivimäki, M.; Singh-Manoux, A.; Sabia, S. The association between accelerometer-assessed physical activity and respiratory function in older adults differs between smokers and non-smokers. Sci. Rep. 2019, 9, 10270. [Google Scholar] [CrossRef]
  49. Carsin, A.E.; Keidel, D.; Fuertes, E.; Imboden, M.; Weyler, J.; Nowak, D.; Heinrich, J.; Erquicia, S.P.; Martinez-Moratalla, J.; Huerta, I.; et al. Regular Physical Activity Levels and Incidence of Restrictive Spirometry Pattern: A Longitudinal Analysis of 2 Population-Based Cohorts. Am. J. Epidemiol. 2020, 189, 1521–1528. [Google Scholar] [CrossRef]
  50. Spruit, M.A.; Pitta, F.; McAuley, E.; ZuWallack, R.L.; Nici, L. Pulmonary Rehabilitation and Physical Activity in Patients with Chronic Obstructive Pulmonary Disease. Am. J. Respir. Crit. Care Med. 2015, 192, 924–933. [Google Scholar] [CrossRef]
  51. Arem, H.; Moore, S.C.; Patel, A.; Hartge, P.; Berrington de Gonzalez, A.; Visvanathan, K.; Campbell, P.T.; Freedman, M.; Weiderpass, E.; Adami, H.O.; et al. Leisure time physical activity and mortality: A detailed pooled analysis of the dose-response relationship. JAMA Intern. Med. 2015, 175, 959–967. [Google Scholar] [CrossRef]
  52. Ekelund, U.; Sanchez-Lastra, M.A.; Dalene, K.E.; Tarp, J. Dose-response associations, physical activity intensity and mortality risk: A narrative review. J. Sport. Health Sci. 2024, 13, 24–29. [Google Scholar] [CrossRef]
  53. Longobardi, I.; Prado, D.; Goessler, K.F.; Meletti, M.M.; de Oliveira Júnior, G.N.; de Andrade, D.C.O.; Gualano, B.; Roschel, H. Oxygen uptake kinetics and chronotropic responses to exercise are impaired in survivors of severe COVID-19. Am. J. Physiol. Heart Circ. Physiol. 2022, 323, H569–H576. [Google Scholar] [CrossRef] [PubMed]
  54. Burn, N.L.; Weston, M.; Maguire, N.; Atkinson, G.; Weston, K.L. Effects of Workplace-Based Physical Activity Interventions on Cardiorespiratory Fitness: A Systematic Review and Meta-Analysis of Controlled Trials. Sports Med. 2019, 49, 1255–1274. [Google Scholar] [CrossRef] [PubMed]
  55. Lavie, C.J.; Church, T.S.; Milani, R.V.; Earnest, C.P. Impact of physical activity, cardiorespiratory fitness, and exercise training on markers of inflammation. J. Cardiopulm. Rehabil. Prev. 2011, 31, 137–145. [Google Scholar] [CrossRef]
  56. Fry, A.; Littlejohns, T.J.; Sudlow, C.; Doherty, N.; Adamska, L.; Sprosen, T.; Collins, R.; Allen, N.E. Comparison of Sociodemographic and Health-Related Characteristics of UK Biobank Participants With Those of the General Population. Am. J. Epidemiol. 2017, 186, 1026–1034. [Google Scholar] [CrossRef]
  57. Brigham, E.P.; West, N.E. Diagnosis of asthma: Diagnostic testing. Int. Forum Allergy Rhinol. 2015, 5 (Suppl. S1), S27–S30. [Google Scholar] [CrossRef]
  58. Gao, M.; Jebb, S.A.; Aveyard, P.; Ambrosini, G.L.; Perez-Cornago, A.; Carter, J.; Sun, X.; Piernas, C. Associations between dietary patterns and the incidence of total and fatal cardiovascular disease and all-cause mortality in 116,806 individuals from the UK Biobank: A prospective cohort study. BMC Med. 2021, 19, 83. [Google Scholar] [CrossRef]
  59. Chudasama, Y.V.; Khunti, K.K.; Zaccardi, F.; Rowlands, A.V.; Yates, T.; Gillies, C.L.; Davies, M.J.; Dhalwani, N.N. Physical activity, multimorbidity, and life expectancy: A UK Biobank longitudinal study. BMC Med. 2019, 17, 108. [Google Scholar] [CrossRef] [PubMed]
  60. Zemedikun, D.T.; Gray, L.J.; Khunti, K.; Davies, M.J.; Dhalwani, N.N. Patterns of Multimorbidity in Middle-Aged and Older Adults: An Analysis of the UK Biobank Data. Mayo Clin. Proc. 2018, 93, 857–866. [Google Scholar] [CrossRef]
  61. Barnett, K.; Mercer, S.W.; Norbury, M.; Watt, G.; Wyke, S.; Guthrie, B. Epidemiology of multimorbidity and implications for health care, research, and medical education: A cross-sectional study. Lancet 2012, 380, 37–43. [Google Scholar] [CrossRef]
  62. Chu, H.; Nie, L.; Cole, S.R. Estimating the relative excess risk due to interaction: A bayesian approach. Epidemiology 2011, 22, 242–248. [Google Scholar] [CrossRef]
  63. Sun, S.; Cao, W.; Qiu, H.; Ran, J.; Lin, H.; Shen, C.; Siu-Yin Lee, R.; Tian, L. Benefits of Physical Activity Not Affected by Air Pollution: A Prospective Cohort Study. Int. J. Epidemiol. 2020, 49, 142–152. [Google Scholar] [CrossRef] [PubMed]
Greenhealth 01 00006 g001
Figure 1. Associations between total habitual physical activity volume and all-cause mortality risk among people with different levels of airflow obstruction. (A). Dose–response associations of total physical activity volume and all-cause mortality risks; (B). All-cause mortality risk reduction percentage due to engaging in physical activity of 600, 1200, 1800, 2400, 3000, and 3600 MET-min/wk, respectively. Results were estimated by multivariable-adjusted Cox proportional hazard regression model, with adjustment for age at recruitment, sex, BMI, TDI, household income, employment status, education level, smoking status, frequency of intaking alcohol, diet level, sedentary time, and secondhand smoke exposure. The knots of restricted cubic splines were placed at the 5th, 35th, 65th, and 95th percentiles to assess the potential non-linearity according to previous studies. The airflow obstruction severity was determined by a modification of the Obstructive Lung Disease (GOLD) criteria: obstructed (FEV1/FVC < 70%), restricted (FEV1/FVC ≥ 70%, and FEV1 < 80% predicted), and normal (FEV1/FVC ≥ 70%, and FEV1 < 80% predicted). And adults with obstructed or restricted airflow were regarded to have declining lung function. Abbreviations: TDI, Townsend deprivation index; MET-min/wk, metabolic equivalent minute per week; HR, hazard ratio.
Figure 1. Associations between total habitual physical activity volume and all-cause mortality risk among people with different levels of airflow obstruction. (A). Dose–response associations of total physical activity volume and all-cause mortality risks; (B). All-cause mortality risk reduction percentage due to engaging in physical activity of 600, 1200, 1800, 2400, 3000, and 3600 MET-min/wk, respectively. Results were estimated by multivariable-adjusted Cox proportional hazard regression model, with adjustment for age at recruitment, sex, BMI, TDI, household income, employment status, education level, smoking status, frequency of intaking alcohol, diet level, sedentary time, and secondhand smoke exposure. The knots of restricted cubic splines were placed at the 5th, 35th, 65th, and 95th percentiles to assess the potential non-linearity according to previous studies. The airflow obstruction severity was determined by a modification of the Obstructive Lung Disease (GOLD) criteria: obstructed (FEV1/FVC < 70%), restricted (FEV1/FVC ≥ 70%, and FEV1 < 80% predicted), and normal (FEV1/FVC ≥ 70%, and FEV1 < 80% predicted). And adults with obstructed or restricted airflow were regarded to have declining lung function. Abbreviations: TDI, Townsend deprivation index; MET-min/wk, metabolic equivalent minute per week; HR, hazard ratio.
Greenhealth 01 00006 g001
Table 1. Baseline information of characteristics among 201,596 participants in the UK Biobank.
Table 1. Baseline information of characteristics among 201,596 participants in the UK Biobank.
Level *OverallAirflow Obstruction Severity †p Value
NormalRestrictedObstructed
Total number201,596161,600 (80.16)13,593 (6.74)26,403 (13.10)
Death <0.001
  Yes10,560 (5.24)7182 (4.44)1080 (7.95)2298 (8.70)
Sex <0.001
  Female113,255 (56.18)93,730 (58.00)7747 (56.99)11,778 (44.61)
  Male88,341 (43.82)67,870 (42.00)5846 (43.01)14,625 (55.39)
Age, year56.38 (7.99)55.85 (7.98)57.00 (7.78)59.25 (7.45)<0.001
Follow-up, year11.81 (1.54)11.84 (1.46)11.67 (1.82)11.67 (1.85)<0.001
BMI level <0.001
  Healthy67,676 (33.57)54,647 (33.82)2984 (21.95)10,045 (38.04)
  Underweight619 (0.31)458 (0.28)25 (0.18)136 (0.52)
  Overweight87,148 (43.23)70,519 (43.64)5379 (39.57)11,250 (42.61)
  Obesity46,153 (22.89)35,976 (22.26)5205 (38.29)4972 (18.83)
TDI <0.001
  Quintile 140,316 (20.00)32,879 (20.35)2477 (18.22)4960 (18.79)
  Quintile 240,319 (20.00)32,667 (20.21)2569 (18.90)5083 (19.25)
  Quintile 340,316 (20.00)32,518 (20.12)2571 (18.91)5227 (19.80)
  Quintile 440,325 (20.00)32,272 (19.97)2765 (20.34)5288 (20.03)
  Quintile 540,320 (20.00)31,264 (19.35)3211 (23.62)5845 (22.14)
Yearly household income, £ <0.001
  Less than 18,00032,140 (15.94)23,845 (14.76)2729 (20.08)5566 (21.08)
  18,000 to 30,99944,101 (21.88)34,562 (21.39)3198 (23.53)6341 (24.02)
  31,000 to 51,99949,915 (24.76)41,055 (25.41)3107 (22.86)5753 (21.79)
  52,000 to 100,00042,807 (21.23)35,782 (22.14)2463 (18.12)4562 (17.28)
  Greater than 100,00012,176 (6.04)10,305 (6.38)586 (4.31)1285 (4.87)
  Unknown20,457 (10.15)16,051 (9.93)1510 (11.11)2896 (10.97)
Employment status <0.001
  Paid121,421 (60.23)100,689 (62.31)7737 (56.92)12,995 (49.22)
  Retired66,778 (33.12)50,300 (31.13)4693 (34.53)11,785 (44.64)
  Unpaid13,397 (6.65)10,611 (6.57)1163 (8.56)1623 (6.15)
Education level <0.001
  Higher degree105,487 (52.33)86,154 (53.31)6315 (46.46)13,018 (49.31)
  Any school degree59,074 (29.30)48,076 (29.75)4094 (30.12)6904 (26.15)
  Vocational qualifications11,111 (5.51)8493 (5.26)880 (6.47)1738 (6.58)
  Other25,924 (12.86)18,877 (11.68)2304 (16.95)4743 (17.96)
Alcohol intake frequency <0.001
  Never11,203 (5.56)8629 (5.34)1025 (7.54)1549 (5.87)
  Occasional40,763 (20.22)32,812 (20.30)3220 (23.69)4731 (17.92)
  Moderate105,155 (52.16)85,262 (52.76)6636 (48.82)13,257 (50.21)
  Heavy44,475 (22.06)34,897 (21.59)2712 (19.95)6866 (26.00)
Smoking Status <0.001
  Never115,846 (57.46)95,938 (59.37)7443 (54.76)12,465 (47.21)
  Previous79,954 (39.66)61,296 (37.93)5730 (42.15)12,928 (48.96)
  Current5796 (2.88)4366 (2.70)420 (3.09)1010 (3.83)
Diet <0.001
  Unhealthy25,396 (12.60)20,176 (12.49)1927 (14.18)3293 (12.47)
  Moderate57,461 (28.50)45,625 (28.23)4102 (30.18)7734 (29.29)
  Healthy115,670 (57.38)93,483 (57.85)7296 (53.67)14,891 (56.40)
  Unknown3069 (1.52)2316 (1.43)268 (1.97)485 (1.84)
Sedentary time <0.001
  Short71,089 (35.26)57,950 (35.86)4129 (30.38)9010 (34.12)
  Moderate64,717 (32.10)51,932 (32.14)4358 (32.06)8427 (31.92)
  Long64,094 (31.79)50,460 (31.23)4953 (36.44)8681 (32.88)
  Unknown1696 (0.84)1258 (0.78)153 (1.13)285 (1.08)
Secondhand smoke exposure <0.001
  No150,836 (74.82)121,729 (75.33)9664 (71.10)19,443 (73.64)
  Yes39,997 (19.84)31,423 (19.44)3096 (22.78)5478 (20.75)
  Unknown10,763 (5.34)8448 (5.23)833 (6.13)1482 (5.61)
FEV1, liter2.86 (0.74)2.99 (0.70)2.08 (0.47)2.46 (0.73)<0.001
FVC, liter 3.76 (0.94)3.84 (0.89)2.75 (0.64)3.77 (1.05)<0.001
FEV1/FVC percentage, %76.03 (5.94)77.85 (3.89)75.72 (3.75)65.05 (5.35)<0.001
Physical activity volume, MET-min/wk
  Total2532.15 (2314.96)2541.28 (2310.57)2285.50 (2269.02)2603.25 (2356.91)<0.001
  Low intensity1016.66 (1048.56)1019.70 (1049.88)965.89 (1046.75)1024.17 (1040.67)<0.001
  Moderate intensity900.36 (1161.33)897.74 (1156.15)824.67 (1135.52)955.39 (1202.83)<0.001
  Vigorous intensity615.13 (931.323)623.84 (930.05)494.94 (861.471)623.69 (968.73)<0.001
Level Physical activity volume <0.001
  Low36,058 (17.89)28,245 (17.48)3251 (23.92)4562 (17.28)
  Moderate83,500 (41.42)67,072 (41.50)5559 (40.90)10,869 (41.17)
  High82,038 (40.69)66,283 (41.02)4783 (35.19)10,972 (41.56)
*: The statistics are shown as mean [standard deviation (SD)] for continuous variables and frequency (%) for categorical variables. †: Physical activity levels were defined according to a standard scoring criteria of International Physical Activity Questionnaire (IPAQ): low (<600 MET-min/week), moderate (600 to 3000 MET-min/week), and high (≥3000 MET-min/week). The airflow obstruction severity was determined by a modification of the Obstructive Lung Disease (GOLD) criteria: obstructed (FEV1/FVC < 70%), restricted (FEV1/FVC ≥ 70%, and FEV1 < 80% predicted), and normal (FEV1/FVC ≥ 70%, and FEV1 ≥ 80% predicted). Abbreviations: TDI, Townsend deprivation index; BMI, body mass index; FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity, MET-min/wk, metabolic equivalent minutes per week.
Table 2. Associations of physical activity volume and lung function with all-cause mortality.
Table 2. Associations of physical activity volume and lung function with all-cause mortality.
Variable Case/No. (%)Model 1 *Model 2 *Model 3 *
Airflow obstruction severity †
  Normal7182/161,600 (4.44)1.0001.0001.000
  Restricted1080/13,593 (7.95)1.671 (1.567, 1.781)1.458 (1.367, 1.556)1.439 (1.348, 1.535)
  Obstructed2298/26,403 (8.70)1.415 (1.35, 1.484)1.351 (1.287, 1.417)1.346 (1.282, 1.412)
  Pfor trend-<0.001<0.001<0.001
Physical activity †
  Continuous, MET-min/week 10,560/201,596 (5.24)0.996 (0.995, 0.997)0.996 (0.996, 0.997)0.997 (0.996, 0.997)
  Low2282/36,058 (6.33)1.0001.0001.000
  Moderate4317/83,500 (5.17)0.767 (0.729, 0.807)0.815 (0.775, 0.858)0.824 (0.782, 0.867)
  High3961/82,038 (4.83)0.708 (0.672, 0.745)0.756 (0.717, 0.797)0.767 (0.727, 0.808)
  Pfor trend-<0.001<0.001<0.001
*: Model 1 was adjusted for sex and age at recruitment. Model 2 was further adjusted for socioeconomic (household income, Townsend deprivation index, employment status, and education) and lifestyle factors (diet level, smoking status, alcohol intake frequency, sedentary time, body mass index (BMI), and secondhand smoke exposure). Model 3 was mutually adjusted. Results were presented as HR and corresponding 95% confidence intervals (CI). †: The volume of physical activity levels was defined according to standard scoring criteria of International Physical Activity Questionnaire (IPAQ): low (<600 MET-min/week), moderate (600 to 3000 MET-min/week), and high (≥3000 MET-min/week). The airflow obstruction severity was determined by a modification of the Obstructive Lung Disease (GOLD) criteria: obstructed (FEV1/FVC < 70%), restricted (FEV1/FVC ≥ 70%, and FEV1 < 80% predicted), and normal (FEV1/FVC ≥ 70%, and FEV1 ≥ 80% predicted). Abbreviations: No., number; MET, metabolic equivalent; min, minutes; FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity.
Table 3. Subgroup and joint analyses of the associations between physical activity volume, airflow obstruction severity, and all-cause mortality.
Table 3. Subgroup and joint analyses of the associations between physical activity volume, airflow obstruction severity, and all-cause mortality.
Airflow Obstruction Severity †Level PA Volume *,†RERI ‡p for Interaction §
LowModerateHighModerate PA LevelHigh PA Level
Subgroup analysis-
  Normal1.0000.871 (0.817, 0.928)0.822 (0.769, 0.878)--
  Restricted1.0000.727 (0.629, 0.840)0.650 (0.556, 0.761)--
  Obstructed1.0000.771 (0.692, 0.859)0.701 (0.626, 0.784)--
Joint analysis0.010
  Normal1.0000.870 (0.816, 0.927)0.818 (0.767, 0.873)--
  Restricted1.679 (1.492, 1.889)1.214 (1.088, 1.354)1.072 (0.949, 1.212)−0.34 (−0.56, −0.11)−0.43 (−0.65, −0.20)
  Obstructed1.511 (1.366, 1.672)1.150 (1.057, 1.251)1.053 (0.966, 1.149)−0.23 (−0.40, −0.07)−0.28 (−0.44, −0.11)
*: Results were estimated by multivariable-adjusted Cox regression model with adjustment for age at recruitment (per 10-year interval), sex (male and female), BMI (underweight, normal, overweight, and obese), TDI (continuous), household income (less than 18,000 £, 18,000–30,999 £, 31,000–51,999 £, 52,000–100,000 £, greater than 100,000 £, and unknown), employment status (paid, retired, and unpaid), education level (higher degree, any school degree, vocational qualifications, and other), smoking status (never, ever, and current smoker), frequency of intaking alcohol (never, occasional, moderate, and heavy), diet level (unhealthy, moderate, healthy, and unknown), sedentary time (short, moderate, long, and unknown), and secondhand smoke exposure (no, yes, and unknown). Results were estimated as hazard ratios and their corresponding 95% confidence intervals (HR, 95% CI). †: Physical activity levels were defined according to standard scoring criteria of International Physical Activity Questionnaire (IPAQ): low (<600 MET-min/wk), moderate (600 to 3000 MET-min/wk), and high (≥3000 MET-min/wk). The airflow obstruction severity was determined by a modification of the Obstructive Lung Disease (GOLD) criteria: obstructed (FEV1/FVC < 70%), restricted (FEV1/FVC ≥ 70%, and FEV1 < 80% predicted), and normal (FEV1/FVC ≥ 70%, and FEV1 ≥ 80% predicted). ‡: The estimates of RERI were calculated based on the reference group with the worst lung function (obstructed) and low PA level. §: Likelihood tests were applied to examine the significance of multiplicative interaction by comparing the model with and without the interaction term. p < 0.05 of the product term indicates a multiplicative interaction. Abbreviations: PA, physical activity; FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity; RERI, relative excess risk due to interaction; MET-min/wk, metabolic equivalent minutes per week.
Table 4. Associations between weekly duration of LMPA, VPA, and all-cause mortality among subjects with different airflow obstruction severity.
Table 4. Associations between weekly duration of LMPA, VPA, and all-cause mortality among subjects with different airflow obstruction severity.
PA AttributesAirflow Obstruction Severity *,†
NormalRestrictedObstructed
Case/No. (%)HR (95% CI)Case/No. (%)HR (95% CI)Case/No. (%)HR (95% CI)
LMPA duration, min/wk
0–45458/8471 (5.41)1.000145/1122 (12.92) 183/1350 (13.56)
46–1491028/24,297 (4.23)0.796 (0.713, 0.889)196/2358 (8.31)0.729 (0.586, 0.906)328/3774 (8.69)0.699 (0.583, 0.839)
150–3001660/39,443 (4.21)0.762 (0.687, 0.846)242/3267 (7.40)0.654 (0.53, 0.807)529/6196 (8.54)0.677 (0.571, 0.802)
301–4501048/24,095 (4.35)0.749 (0.67, 0.838)136/1789 (7.60)0.626 (0.492, 0.795)344/4048 (8.50)0.644 (0.537, 0.773)
451–600760/16,669 (4.56)0.771 (0.685, 0.867)94/1210 (7.77)0.636 (0.487, 0.829)193/2706 (7.13)0.547 (0.445, 0.671)
601–900888/19,161 (4.63)0.729 (0.65, 0.818)119/1516 (7.85)0.608 (0.474, 0.779)282/3284 (8.59)0.594 (0.492, 0.719)
901–1200489/10,348 (4.73)0.724 (0.637, 0.824)59/825 (7.15)0.55 (0.404, 0.748)178/1827 (9.74)0.656 (0.532, 0.809)
>1200851/19,116 (4.45)0.699 (0.623, 0.785)89/1506 (5.91)0.473 (0.361, 0.621)261/3218 (8.11)0.57 (0.47, 0.691)
VPA duration, min/wk
0–93202/60,685 (5.28)1.000667/6449 (10.34) 1203/10,529 (11.43)
10–741846/45,018 (4.10)0.875 (0.826, 0.928)196/3481 (5.63)0.668 (0.568, 0.786)466/7037 (6.62)0.682 (0.612, 0.761)
75–1501066/28,054 (3.80)0.867 (0.808, 0.93)114/1907 (5.98)0.718 (0.587, 0.878)287/4247 (6.76)0.738 (0.648, 0.841)
151–225451/12,055 (3.74)0.881 (0.797, 0.973)33/736 (4.48)0.545 (0.384, 0.776)134/1885 (7.11)0.814 (0.679, 0.974)
226–300293/8487 (3.45)0.796 (0.706, 0.898)39/522 (7.47)0.926 (0.669, 1.282)91/1407 (6.47)0.716 (0.578, 0.887)
>300324/7301 (4.44)0.866 (0.771, 0.971)31/498 (6.22)0.668 (0.465, 0.96)117/1298 (9.01)0.829 (0.685, 1.004)
VPA proportion, %
03202/60,685 (5.28)1.000667/6449 (10.34)1.0001203/10,529 (11.43)1.000
0–241748/41,162 (4.25)0.859 (0.81, 0.911)187/3067 (6.10)0.668 (0.465, 0.96)480/6659 (7.21)0.705 (0.633, 0.786)
25–491422/36,849 (3.86)0.861 (0.808, 0.917)146/2553 (5.72)0.691 (0.586, 0.816)403/5722 (7.04)0.763 (0.68, 0.856)
50–74670/18,774 (3.57)0.881 (0.809, 0.958)64/1259 (5.08)0.684 (0.571, 0.821)184/2875 (6.40)0.735 (0.628, 0.86)
≥75140/4130 (3.39)0.976 (0.823, 1.157)16/265 (6.04)0.641 (0.495, 0.83)28/618 (4.53)0.606 (0.416, 0.883)
*: All results were calculated multivariable-adjusted Cox proportional hazard regression models, with adjustment for age at recruitment, sex, BMI, TDI, household income, employment status, education level, smoking status, frequency of intaking alcohol, diet level, sedentary time, and secondhand smoke exposure. Because a small number of participants reported no LMPA over the entire follow-up, the reference group for LMPA was broadened to 0–45 min/wk. VPA proportion is the proportion of VPA volume to total PA volume to indicate participants’ intensity level. †: The airflow obstruction severity was determined by a modification of the Obstructive Lung Disease (GOLD) criteria: obstructed (FEV1/FVC < 70%), restricted (FEV1/FVC ≥ 70%, and FEV1 < 80% predicted), and normal (FEV1/FVC ≥ 70%, and FEV1 < 80% predicted). And adults with obstructed or restricted airflow were regarded to have impaired lung function. Abbreviations: No., number; PA, physical activity; LMPA, low-to-moderate intensity of PA; VPA, vigorous intensity of PA; HR, hazard ratio; CI, confidence interval.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Chen, L.; Wang, C.; Zhang, S.; Wei, S.; Zhao, J.; Zhang, Z. Habitual Physical Activity and All-Cause Mortality Among Individuals with and Without Impaired Lung Function: Findings from a Prospective Cohort Study. Green Health 2025, 1, 6. https://doi.org/10.3390/greenhealth1020006

AMA Style

Chen L, Wang C, Zhang S, Wei S, Zhao J, Zhang Z. Habitual Physical Activity and All-Cause Mortality Among Individuals with and Without Impaired Lung Function: Findings from a Prospective Cohort Study. Green Health. 2025; 1(2):6. https://doi.org/10.3390/greenhealth1020006

Chicago/Turabian Style

Chen, Lan, Chongjian Wang, Shiyu Zhang, Shengtao Wei, Jinde Zhao, and Zilong Zhang. 2025. "Habitual Physical Activity and All-Cause Mortality Among Individuals with and Without Impaired Lung Function: Findings from a Prospective Cohort Study" Green Health 1, no. 2: 6. https://doi.org/10.3390/greenhealth1020006

APA Style

Chen, L., Wang, C., Zhang, S., Wei, S., Zhao, J., & Zhang, Z. (2025). Habitual Physical Activity and All-Cause Mortality Among Individuals with and Without Impaired Lung Function: Findings from a Prospective Cohort Study. Green Health, 1(2), 6. https://doi.org/10.3390/greenhealth1020006

Article Metrics

Back to TopTop