Factors contributing to differences in physical activity levels in (pre)frail older adults living in rural areas of China

Xin Zhang; Xiaoping Zheng; Hans Hobbelen; Barbara van Munster; Qian Tong; Tianzhuo Yu; Feng Li; Claudine JC Lamoth

doi:10.1371/journal.pone.0335607

. 2025 Nov 4;20(11):e0335607. doi: 10.1371/journal.pone.0335607

Factors contributing to differences in physical activity levels in (pre)frail older adults living in rural areas of China

Xin Zhang ^1,^2,^*, Xiaoping Zheng ³, Hans Hobbelen ^4,⁵, Barbara van Munster ⁶, Qian Tong ⁷, Tianzhuo Yu ², Feng Li ², Claudine JC Lamoth ¹

Editor: Hidetaka Hamasaki⁸

PMCID: PMC12585090 PMID: 41187200

Abstract

Introduction

Physical Activity (PA) is essential for enhancing the physical function of pre-frail and frail older adults. However, among this group, PA-levels vary significantly. Identifying the factors contributing to these differences could support tailored PA interventions. This study aims to examine factors associated with physical activity levels among pre-frail and frail older adults in rural China.

Methods

This is a cross-sectional study. A total of 284 (pre)frail older adults (aged ≥60 years) were included from ten rural healthcare centers in Northeast China. Participants were categorized into low-moderate and high physical activity groups assessed using the Short Form International Physical Activity Questionnaire. Four-dimensional data were collected, including demographics, health behaviors, objective physical performance measures, and self-reported perceived health profiles. Extreme Gradient Boosting (XGBoost), a machine learning algorithm, was employed for binary classification (low-moderate vs. high physical activity). Model performance was assessed using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, precision, and F1-score. To enhance interpretability, SHapley Additive exPlanations (SHAP) were utilized to identify key predictive variables.

Results

Mean age of participants was 70 years (59% female, 86% farmers). The low-moderate group averaged 1,187 MET/week, while the high physical activity group reached 8,162 MET/week. Physical performance tests showed significantly better scores in the high PA group. The XGBoost model achieved 82.4% accuracy (AUC: 0.769, specificity: 90%, sensitivity: 63%). SHAP analysis revealed that self-reported social support, general health, ambulation, and physical performance measures were the most important factors.

Conclusion

The high physical activity group demonstrated better physical function than the low-moderate physical activity group; though, both groups showed poorer physical function compared to the general older population. Self-reported health perceptions and social support significantly correlated with physical activity levels. Addressing these factors through targeted interventions—including community-based social support programs and structured mobility-enhancing exercises—may contribute to improved health outcomes and enhanced quality of life in this population.

1. Introduction

China’s population is aging rapidly. It is predicted that by 2050, the number of older adults over 65 will rise to 26.1% of the total population in China [1]. As highlighted by a recent national survey in China, over 70% of older adults live in rural areas [2], with rural regions experiencing a disproportionately high burden of aging due to urban migration of younger populations. Compared to older individuals living in urban areas, older adults in rural China often face limited healthcare resources, lower socioeconomic status, and increased vulnerability to age-related health conditions [3,4].

Among older adults, one of the most common clinical syndromes is frailty, especially in northern China [4–6]. Frailty is a multicomponent condition, associated with increased dependency and a heightened risk of mortality [7]. Fried Phenotype is one of the most widely used frailty frameworks to assess physical frailty and operationalizes frailty based on five criteria: unintentional weight loss, muscle weakness (low grip strength), exhaustion, slow walking speed, and low physical activity [8].

Research indicates significant disparities in physical activity levels between rural and urban environments, with older individuals in rural areas exhibiting higher levels of physical activity. The international collaborative study of cardiovascular disease in Asia found that among individuals aged 35–75, 78.1% of rural residents engaged in at least 30 minutes of moderate-to-vigorous physical activity daily, compared to just 21.8% of urban residents [9]. Similar trends have been observed in other studies. A cross-sectional study in the Tarragona province of Spain reported that older adults in rural areas engaged in higher-intensity physical activities (76%) compared to those in urban areas (65%) [10]. Likewise, research in Portugal found that rural older residents scored higher in health-related quality of life and functional fitness than their urban counterparts [11]. These disparities can be attributed to various factors, including geographic characteristics, urbanization, mechanization of daily tasks, limited free time, and increased reliance on vehicles in urban settings.

Engaging in moderate to vigorous physical activity has shown benefits for (pre)frail older adults, improving body functions and potentially influencing other components of frailty [2,12–14]. Physical decline is just one component of frailty. Frailty is a multidimensional condition influenced by multiple interrelated domains, including nutrition, cognition, psychological well-being, self-perceived health and social support [15,16]. Physical activity can serve as a crucial mediator in preventing (pre)frailty by positively influencing the above interrelated domains. As indicated by the Biopsychosocial Model of Health, both objective health measures and self-perceived health significantly influence physical activity [17]. While objective factors such as grip strength and walking speed establish physical capacity, self-perceived health including self-rated health, self-efficacy, and social support, play a critical role in motivating and sustaining activity [18,19]. A growing body of evidence supports the integration of these dimensions to better understand physical activity behavior within a holistic biopsychosocial framework, especially in understudied populations [20,21]. Self-reported factors, reflecting individuals’ perceptions and beliefs about their health and well-being, have shown strong associations with physical activity levels. Higher self-perceived quality of life and greater social support have been linked to increased levels of physical activity [12,13,16,22,23]. While studies have addressed relationships between various individual health factors and physical activity among older adults using regression approaches, the integration of biological, psychological, and social domains and assessment of interactions among these factors will provide a more comprehensive picture of the relation between physical activity and different health domains.

This study aims to get insight into the interaction of different health domains on physical activity levels in older adults living in rural areas of China who have been identified as physically (pre)frail. Based on the biopsychosocial model of health, we hypothesize that self-perceived health measures will demonstrate associations with physical activity levels among pre-frail and frail older adults, and that the interaction between psychological, biological, and social factors will explain a significantly greater proportion of variance in physical activity than any single domain.

2. Methods

2.1 Study design

The study has been conducted in accordance with the principles expressed in the Declaration of Helsinki and approved by the Human Research Ethics Committee of the School of Nursing, Jilin University (Project-ID 2022082905).

In this observational study, potential participants were recruited from the rural Primary Healthcare Centre (PHC) by 10 local general practitioners from October 1 to November 15 of 2022. The general practitioner introduced the project to the potential participants in advance, then, scheduled an appointment with those who agreed to participate and signed the informed consent form. Considering the seasonal farming and inconvenience of transportation in the rural area, measurements of participants in the same village were scheduled for the same day and conducted at the local PHC. The research group has a long-term collaboration with these rural PHC, and a consensus on recruitment criteria was reached before the start of the project. Considering the low education level of most participants, all questionnaires and self-reported measurements were conducted in a one-on-one question-and-answer format and recorded by the investigators. Investigators were licensed health practitioners and trained to interview participants before the measurements.

2.2 Population

A total of 367 participants older than 60 have been recruited from 10 villages in rural areas of northeast China. Inclusion criteria for participation were, frailty score >0 according to the Fried Phenotype [8]; be able to walk independently without or with a walking stick for at least 3 meters; be able to understand Mandarin; capacity of consent. Exclusion criteria were, pain and severe mobility disability; unstable or persistent serious cardiovascular or respiratory system diseases (e.g., stage IV heart failure and COPD); Inability to communicate (e.g., severe hearing, vision, cognitive and language judged by primary health care practitioner). All participants provided a signed written informed consent before inclusion.

2.3 Assessment instruments

Physical activity (PA) level, was examined by the Short Form International Physical Activity Questionnaire (IPAQ-SF, Chinese version), which collects information on the frequency and duration (i.e., number of days and average time per day) spent in vigorous physical activity, moderate physical activity, walking activity, and sedentary behavior in the last seven consecutive days [24].Participants were labelled as having low to moderate and high physical activity intensities based on the MET (Metabolic equivalent – multiples of the resting metabolic rate) calculated according to the guidelines for interpreting the IPAQ-SF. High physical activity intensity is defined as at least 3 days achieving a minimum of 1,500 MET-minutes/week OR 7 or more days of any combination of walking, moderate-intensity or vigorous-intensity activities achieving a minimum of 3,000 MET-minutes/week. MET value is calculated as 8 × high physical activity (min/week) + 4 × moderate physical activity (min./week) + 3.3 × walking (min./week) [25].

Four groups of variables were collected, demographic characteristics, health behavior, objective health performance, and perceived health profile (Fig 1a), and for details of the assessment instruments refer to the Supporting Information (Supplementary Material-Appendixes).

Fig 1 — (a) Data formulation; (b) Data splitting; (c) Determining optimal hyperparameters based on 5-fold cross validation; (d) XGBoost model training; (e) Model performance evaluation; f. Feature importance analysis.

Demographic and personal characteristics were obtained from the general practitioners directly, including age, Waist Hip Rate (WHR), Body Mass Index (BMI), gender, occupation, education, income, marital status, living status, and cognitive impairment.

Health behavior variables were collected by simplified “yes” or “no” questions, including smoking, drinking, daily intake of fibre, and daily intake of protein.

Objective health performance variables were calculated based on related physical tests or recorded from medical health records, including grip strength (GP), walking speed (WS) (3-meter walking test), time of the Time Up & Go test (TUG), total score of Short Physical Performance Battery (SPPB) [26], Mini-Mental State Examination (MMSE) [27], and number of comorbidities as recorded in the health records.

Regarding perceived health profile variables, the score of the General Anxiety Disorder-7 (GAD-7) [28], Patient Health Questionnaire-9 (PHQ-9) [29], and Social Support Rating Scale (SSRS) were obtained. The SSRS evaluates three key areas of social support: objective support, the actual support received (e.g., financial or material help), subjective support – how much emotional and psychological support a person feels they receive from others. Finally, it examines support utilization, which reflects how well and how frequently an individual makes use of their available support networks [30]. Subjective health and perceived functional strength and ambulation were assessed by three questions: “In general, how would you rate your health?”, “Do you have any difficulty climbing a one floor stair or a 10-meter slope without resting and without assistance?”, “Do you have any difficulty walking 500 meters alone without assistance?”.

2.4 Data analysis

Statistical analysis of differences of individual variables between two groups was performed with Chi-Square tests for categorical variables and Mann-Whitney U tests for physical assessments, as the data were not normally distributed, p < 0.05 was considered statistically significant (Fig 2).

Fig 2 — Dots represent individual subjects, the boxplots depict the median (central horizontal line) and interquartile range (box), while the mean is indicated by a diamond symbol. The right portion of each boxplot displays the full data distribution for the respective group. PA = Physical Activity; MET = Metabolic Equivalent of Task; SPPB = Short Physical Performance Batter; TUG = Timed Up & Go test.

XGBoost (eXtreme Gradient Boosting) was used for this study due to its ability to handle imbalanced data, capture complex non-linear relationships, and provide feature importance insights, making it well-suited for binary classification tasks [31–33]. XGBoost can handle continuous, categorical, and binary variables without normalization because it uses decision trees, which split data based on relative ordering rather than absolute magnitude. All analysis were performed in R-studio (R 4.1.0 `xgboost` package version 1.4.1.1). The dependent variable (label) physical activity intensity was binarized to class 0 for low to moderate physical intensity and class 1 for high physical activity intensity level. Independent variables (input features were the 26 variables from the assessments of which 15 were binary scored variables (yes/no)).

2.4.1 Architecture and pipeline of XGBoost.

The dataset was split into training (80%) and testing sets using stratified sampling to maintain balanced representation of the binary physical activity intensity classes. The data of the remaining 20% of participants were used as the testing dataset (Fig 1b). To address class imbalance (training data set N = 68 class 0 vs N = 160 class 1), class weights were calculated based on the ratio of negative to positive cases in the training dataset. This weight was used to balance the loss function during model training [34]. In this study, the high physical activity intensity group (class 1) was considered as the positive case and the low-moderate physical activity class the negative case.

A hyperparameter tuning process was conducted to optimize XGBoost using the training dataset. A grid search approach was employed to explore potential hyperparameter combinations, including learning rate (eta), maximum tree depth, number of boosting rounds, gamma, column sample by tree, minimum child weight, and subsampling ratio. Details has been provided in the supporting information (Supplementary Material-Appendixes).

To prevent overfitting and enhance model generalizability, a 5-fold cross-validation was applied during the validation phase. The training dataset was split into 5 folds, where each fold was used as validation while the remaining four folds were used to train the model with the selected hyperparameter combinations (Fig 1c). Mean squared error (MSE) was used to assess model performance during validation, and the average MSE across the 5 iterations represented the overall performance of selected hyperparameter combination. After identifying the best hyperparameters, the final model was retrained using the entire training dataset. Subsequently, the testing dataset was used to evaluate the generalizability of the model (Fig 1d).

2.4.2 Model performance assessment.

The overall performance of the XGBoost model was evaluated by the area under the receiver operating characteristic (ROC) curve (AUC). In addition, accuracy, sensitivity, specificity, precision, F1-score, were calculated (Fig 1e). Correct predictions for class 1 (high physical activity intensity group) are true positives (TP) and for class 0 (low to moderate physical activity intensity group) are true negatives (TN), while incorrect predictions are false positives (FP) and false negatives (FN). Accuracy is the proportion of all correct predictions: (TP + TN)/ (TP + FP + TN + FN). Sensitivity represents the proportion of positive cases correctly identified: TP/ (TP + FN). Specificity refers to the proportion of negative cases correctly identified: TN/ (TN + FP). Precision is the ratio of correctly predicted positive cases to all predicted positive cases: TP/ (TP + FP). The F1-score is the harmonic mean of precision and sensitivity: (2 × precision × sensitivity)/ (precision + sensitivity).

2.4.3 Feature importance.

To get insights into the contribution of individual features (variables) to the classification model, SHAP (SHapley Additive exPlanations) values were computed (using the R-packages: SHAPforxgboost and shapviz) (Fig 1f). SHAP derived from cooperative game theory, allocate importance to features based on their contributions to predictions [35]. SHAP values were calculated for the test dataset. SHAP calculates the importance of a feature by comparing the model’s predictions with and without that feature, considering all possible orders and combination of feature inclusion to ensure a fair comparison. Higher absolute SHAP values indicate a high contribution to the classification process. Large positive values indicate that a feature drives the prediction toward the positive class, while negative values indicate that the feature pushes toward the negative class. A SHAP Summary Plot is presented displaying feature importance and effect direction.

3. Results

3.1 General characteristics

Of the 367 recruited participants, 284 were included in the analysis based on inclusion and exclusion criteria. 69 participants were not eligible: 47 participants had a Fried score of zero (indicating no physical frailty), 9 had unstable diseases, and 13 participants were missing the TUG test. After inclusion, an additional 14 participants were excluded during data inspection because their reported values fell outside the possible range for the questionnaire or assessment, likely due to data entry errors.

The mean age of participants was 70 ± 5.5 years (range: 60–87 years), with 59% being female. The majority of participants were farmers with low income and educational levels. The low-moderate physical activity group had a mean MET per week of 1,186.78 (SD = 1,012.52) and the high physical activity group of 8,162.47 (SD = 4,050.45) (Fig 2).

3.2 Difference of observed variables between low-moderate and high PA group

Analysis revealed significant differences between low-moderate and high physical activity groups regarding occupation, with a higher proportion of farmers in the high physical activity group. Income levels also differed significantly between groups, with the high physical activity group reporting lower average monthly income. Lower smoking rates were observed in the high physical activity group (Table 1).

Table 1. Demographics and outcome variables of the two groups.

Variables (unit)	Low-Moderate PA	High PA	Statistic
N	84	200	–	–
Age (y)	70.5 (5.6)	69.2 (5.4)	U = 1.07	p = .380
Gender Female vs Male	58%	60%	Fisher	p = .960
Occupation Farmer vs. other	74%	91%	Χ² = 10.65	p = .001
Education < 6 year	79%	79%	Fisher	p = 1.000
Income < 3000 (￥/m)	90%	96%	Χ² = 4.29	p = .038
Married vs other	76%	81%	Fisher	p = .449
Living with others Yes	85%	90%	Fisher	p = .328
Smoking No	45%	33%	Χ² = 4.15	p = .042
Drinking No	77%	79%	Fisher	p = .959
Daily Fiber Yes	82%	87%	Fisher	p = .447
Daily protein Yes	43%	52%	Fisher	p = .230
Comorbidities <=5	82%	92%	Χ² = 5.91	p = .015
Walk >500 m	58%	85%	Χ² = 22.70	p < .001
Can walk stairs (functional strength)	57%	79%	Χ² = 14.19	p < .001
Good general health	35%	65%	Χ² = 21.54	p < .001
No cognitive impairment	54%	69%	Χ² = 5.73	p = .017
WHR	0.92 (0.06)	0.94 (0.85)	U = 1.01	p = .314
BMI	23.40 (4.00)	24.65 (3.50)	U = 2.19	p = .028
GP (kg)	19.64 (8.48)	21.99 (8.51)	U = 1.87	p = .030
WS (m/s)	0.60 (0.21)	0.67 (0.21)	U = 3.28	p = .001
TUG (seconds)	14.98 (7.91)	12.06 (3.25)	U = 2.43	p = .020
SPPB -	6.67 (2.66)	8.20 (2.34)	U = 4.40	p < .001
MMSE	19.67 (4.96)	21.81 (4.90)	U = 3.16	p = .002
PHQ9	8.31 (5.43)	6.67 (4.57)	U = 2.31	p = .021
GAD7	7.31 (5.70)	6.88 (5.35)	U = 0.61	p = .541
SSRS	32.18 (7.71)	36.35 (7.39)	U = 4.12	p < .001

Open in a new tab

PA = Physical Activity; WHR = Waist-Hip Ratio; BMI = Body Mass Index; GP = Grip strength; WS = Walking Speed; TUG = Timed-up-and-Go; SPPB = Short Physical Performance Battery; MMSE = Minimal Mental State Examination; PHQ9 = Patient Health Questionnaire; GAD7 = General Anxiety Disorder Questionnaire Score; SSRS = Social Support Rate Scale; Χ² = Chi-square test; U = Mann Withney U test.

As illustrated in Fig 2 and Table 1, significant between-group differences were observed in measures derived from physical tests. The high physical activity group, demonstrated better performance in walking speed, grip strength, TUG and SPPB sores compared to the low-moderate physical activity group. The distribution of the data showed wider distribution of the high physical activity group at the higher scores in particular for SPPB, walking speed and grip strength.

Self-perceived health and social functioning scores were significantly different between groups with the exception of the GAD-7 score (Fig 3, Table 1), which was highly variable as indicated by the multimodal distribution of data in both groups.

3.3 Performance of XGBoost model

The XGBoost model demonstrated good classification performance with an accuracy of 82.4% (95% CI: 70–91%). The model achieved an AUC of 0.769, sensitivity of 63%, specificity of 90%, precision of 71%, and F1-score of 67%.

Analysis of feature importance using SHAP values (Fig 4) revealed that the SSRS, general health questionnaire, TUG, ambulation capacity, and SPPB were the most influential predictors. Poor self-reported general health was associated with classification into the low-moderate physical activity group, while good health corresponded with the high physical activity group. Similarly, the ability to walk continuously for 500 meters was predictive of classification into the high physical activity group. While the SPPB and SSRS scores showed significant contributions to the model’s predictions, their distributions were more heterogeneous within groups (Fig 1 and Fig 2).

4. Discussion

This study aimed to examine different factors correlated with physical activity levels in (pre)frail older adults living in rural China. The results highlight significant differences between individuals with low-moderate and high physical activity levels, with key determinants including social support, self-reported general health status, and functional mobility. These findings contribute to the growing body of literature on aging, frailty, and physical activity, particularly in rural populations [2,4].

One of the most prominent findings of this study was the strong association between self-reported health, social support, and physical activity engagement. Consistent with previous research individuals with better self-rated health were more likely to engage in high-intensity physical activity [13]. Self-reported health in this study was assessed using a single-item measure: “In general, how would you rate your health?” with response options of “good” or “bad”. Functional health was evaluated based on the ability to walk 500 meters and ascend a 10-meter slope or a one-floor staircase. Poor functional health, has previously been linked to lower positive affect (enthusiasm, pride, interest) but does not necessarily correlate with negative affect (anxiety, distress, anger) [36]. In line with these findings, participants’ GAD-7 scores in this study, which reflect negative affect, were not elevated. However, this represents an association rather than causation. These results suggest that interventions targeting frailty should focus not only on improving physical function, but also incorporate strategies to enhance psychological well-being and promote positive affect.

The role of social support, as measured by the SSRS, further underscores the psychosocial dimensions of frailty. Social networks and perceived emotional support provide motivation, assistance, and opportunities for engagement in physical activity [14], reinforcing the necessity of integrating social interventions into physical activity programs for (pre)frail older adults. The overlap between groups further indicates that social support alone is rarely sufficient to ensure high levels of physical activity. Success of interventions targeting social support depends on simultaneously addressing other individual specific barriers, such as functional limitations or perceived health status.

All participants in this study were classified as physically pre-frail and frail according to the Fried frailty criteria, with the majority exhibiting slow gait speed (99.3%) and low grip strength (59.2%). Mobility-related physical performance measures were significantly lower in the low-to-moderate physical activity group compared to the high physical activity group. However, normative data provided by Mayhew et al. indicates that even participants in the high physical activity group fell within the lowest 5th percentile for gait speed, grip strength, and TUG performance relative to age-matched counterparts [37]. The observed average gait speed of 0.6 m/sec is below the commonly reported cut-off of <0.8 m/sec for identifying clinically relevant slowness associated with frailty and sarcopenia [38]. Furthermore, the averaged SPPB scores ≤9, confirmed significant decline in walking ability, balance and muscle strength [39]. The IPAQ-SF assesses total physical activity without distinguishing between leisure-time physical activity (LTPA) and occupational physical activity (OPA), which may partly explain the observed paradox of high self-reported PA alongside poor physical performance. These findings addressed the necessity of considering the distinct roles of LTPA and OPA separately in future research, particularly in agricultural populations where occupational demands dominate daily activity patterns.

Despite these physical limitations, participants in the high physical activity group engaged in vigorous-intensity activity on at least three days per week, accumulating a minimum of 1,500 MET-minutes/week, or participated in a combination of walking, moderate-intensity, and vigorous-intensity activities for at least seven days per week, reaching a minimum of 3,000 MET-minutes/week. This may be attributed to the fact that the majority of participants were farmers, indicating physical activity as an integral part of their daily routines.

The XGBoost model achieved a satisfactory classification accuracy of 82.4%. However, the model’s accuracy suggests that unmeasured factors may also contribute to physical activity behaviors in this population. While the model exhibited high specificity (90%), moderate sensitivity (63%) indicates potential misclassification, particularly among individuals near the threshold between activity levels. Therefore, this model is positioned as a screening tool, rather than diagnostic, enabling healthcare providers to efficiently identify a high-risk subgroup for physical inactivity and prioritize limited resources toward those most in need. Future research should prioritize improving sensitivity, through methodological refinements, incorporating additional behavioral and environmental variables, to maintain acceptable specificity.

SHAP values revealed not only the importance of physical function (SPPB, TUG) and the ability to walk 500 meters unassisted, in facilitating greater participation in physical activity, but also highlighted the importance of the social support. A broader perspective on mobility and frailty—beyond the focus on physical frailty and physical performance—could offer deeper insights into the determinants of physical activity.

This perspective could incorporate the role of geographical life-space in shaping mobility behaviors. Life-space refers to the spatial extent within which individuals carry out their daily activities, encompassing the home environment, neighbourhood, and broader community interactions [40–42]. Mobility limitations, influenced by factors such as health status, demographic characteristics, and the built environment, significantly affect older adults’ capacity for physical activity. Behaviors, such as pausing for rest during travel or the balance of time spent indoors versus outdoors, serve as indicators of the broader societal and geographical influences on physical activity [42]. In rural areas, where transportation infrastructure is often limited and communities are spatially dispersed, mobility constraints may restrict physical activity opportunities. However, the presence of strong social networks and expansive open landscapes may mitigate these limitations by fostering active lifestyles [41,43]. Examining these behaviors can elucidate personal, social, and environmental determinants of physical activity among frail older adults. Understanding these contextual factors is crucial for developing targeted interventions that promote physical activity and mobility among older adults in rural areas.

It is worth noting that rural environments vary considerably across regions. For instance, Northeast China’s rural areas characterized by severe continental monsoon climates with extended sub-zero winters, creating distinct living conditions compared to southern regions’ milder year-round climate. These geographical variations, along with distinct cultural practices and social structures across different rural communities, may influence the generalizability of our findings. Nevertheless, our findings provide a valuable methodological framework that can inform future comparative studies across diverse rural settings.

The implications of these findings are substantial for interventions targeting frailty and physical inactivity in rural older adults. Given the significant role of social support, community-based programs emphasizing group activities, peer support, and family involvement may enhance participation rates. Considering that most of rural (pre)frail older adults already engaged in sufficient physical activity level, interventions for this population may need to focus on additional aspects, such as nutrition, which could provide greater health benefits [16,44]. For the PA intervention in this population, the primary objective is not to increase total PA volume, but to mitigate the detrimental effects of work-related movement and promote health-enhancing behaviors. Proposed strategies include health education to improve biomechanics and reduce injury risk, enhancing social participation during repetitive labor, and tailored community physical activity program during off-peak seasons designed to counteract physical strain and foster social engagement.

A limitation of this study pertains to the reliance on self-reported questionnaires for assessing physical activity, particularly the IPAQ-SF. This instrument was chosen because of several methodological and practical considerations specific to our study context in rural Northeast China. Although the IPAQ-SF is widely used to assess physical activity in the aging population, a recent systematic review reported that compared to objective physical activity measures with wearable sensors, the IPAQ-SF tends to overestimate activity levels [25]. The results reflect self-reported rather than objectively measured physical activity. While the associations identified between variables remain meaningful within the framework of this study, we recommend that future research incorporate device-based measures to further validate and extend these findings [45].

5. Conclusions

This study provides insights into the factors correlated with physical activity levels in (pre)frail older adults in rural China. The findings highlight the critical role of self-reported general health, social support, and functional mobility in determining physical activity levels. Addressing these factors through targeted interventions, including community-based social support programs and structured mobility-enhancing exercises, may contribute to improved health outcomes and enhanced quality of life in this population. Future research should explore the integration of objective physical activity monitoring and predictive model to further refine intervention strategies and ensure their effectiveness in rural settings.

Supporting information

S1 File. Supplementary material-appendixes.

(DOCX)

pone.0335607.s001.docx^{(26.4KB, docx)}

S2 File. Data XGBOOST.

(CSV)

pone.0335607.s002.csv^{(22.5KB, csv)}

Acknowledgments

We extend our sincere gratitude to all members of the Alpha Care Team for their invaluable contributions to the organizational design during the onsite survey.

Data Availability

All relevant data are within the paper and its Supporting Information files.

Funding Statement

This study was financially supported by Rijksuniversiteit Groningen (https://www.rug.nl) in the form of a grant (P312319) received by XZ. This study was also financially supported by the China Scholarship Council in the form of an award (202206170103) received by XZ.

References

1.Feng Z, Glinskaya E, Chen H, Gong S, Qiu Y, Xu J, et al. Long-term care system for older adults in China: policy landscape, challenges, and future prospects. Lancet. 2020;396(10259):1362–72. doi: 10.1016/S0140-6736(20)32136-X [DOI] [PubMed] [Google Scholar]
2.Deng SY, Zhao IY, Ho M-H, Saravanakumar P, Molassiotis A, Montayre J. Rural-urban disparities in healthy ageing: Evidence from a national study in China. Collegian. 2022;29(6):843–9. doi: 10.1016/j.colegn.2022.06.008 [DOI] [Google Scholar]
3.Li H, Zeng Y, Gan L, Tuersun Y, Yang J, Liu J, et al. Urban-rural disparities in the healthy ageing trajectory in China: a population-based study. BMC Public Health. 2022;22(1):1406. doi: 10.1186/s12889-022-13757-x [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Xu R, Li Q, Guo F, Zhao M, Zhang L. Prevalence and risk factors of frailty among people in rural areas: a systematic review and meta-analysis. BMJ Open. 2021;11(4):e043494. doi: 10.1136/bmjopen-2020-043494 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Zeng X-Z, Meng L-B, Li Y-Y, Jia N, Shi J, Zhang C, et al. Prevalence and factors associated with frailty and pre-frailty in the older adults in China: a national cross-sectional study. Front Public Health. 2023;11:1110648. doi: 10.3389/fpubh.2023.1110648 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Zhou Q, Li Y, Gao Q, Yuan H, Sun L, Xi H, et al. Prevalence of Frailty Among Chinese Community-Dwelling Older Adults: A Systematic Review and Meta-Analysis. Int J Public Health. 2023;68:1605964. doi: 10.3389/ijph.2023.1605964 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Dent E, Martin FC, Bergman H, Woo J, Romero-Ortuno R, Walston JD. Management of frailty: opportunities, challenges, and future directions. Lancet. 2019;394(10206):1376–86. doi: 10.1016/S0140-6736(19)31785-4 [DOI] [PubMed] [Google Scholar]
8.Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci. 2001;56(3):M146-56. doi: 10.1093/gerona/56.3.m146 [DOI] [PubMed] [Google Scholar]
9.Muntner P, Gu D, Wildman RP, Chen J, Qan W, Whelton PK, et al. Prevalence of physical activity among Chinese adults: results from the International Collaborative Study of Cardiovascular Disease in Asia. Am J Public Health. 2005;95(9):1631–6. doi: 10.2105/AJPH.2004.044743 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Jiménez-Ten Hoevel C, Queral J, Besora-Moreno M, Tarro L, Sospedra-Senar A, Tomàs-Piqueras J, et al. Health status, lifestyle and quality of life in older adults of rural and urban areas of Tarragona province (Catalonia, Spain): a cross-sectional descriptive study. BMC Public Health. 2025;25(1):895. doi: 10.1186/s12889-025-22045-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Furtado GE, Vaz C, Bovolini A, Marques E, Serra N, Costa-Brito AR, et al. The Impact of Physical Activity Levels and Functional Fitness Status on the Quality of Life Perceived by Older Adults Living in Rural and Urban Areas: The Portuguese Inland Case. Healthcare (Basel). 2022;10(7):1266. doi: 10.3390/healthcare10071266 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Wang Y, Wang X, Zhu X, Sun Y, Han B, Chen T. Physical activity, physical frailty and depressive symptoms among Chinese male and female older adults: do different indicators make a difference?. Front Public Health. 2023;11:1280952. doi: 10.3389/fpubh.2023.1280952 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Steinhoff P, Reiner A. Physical activity and functional social support in community-dwelling older adults: a scoping review. BMC Public Health. 2024;24(1):1355. doi: 10.1186/s12889-024-18863-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Bauman AE, Reis RS, Sallis JF, Wells JC, Loos RJF, Martin BW, et al. Correlates of physical activity: why are some people physically active and others not?. Lancet. 2012;380(9838):258–71. doi: 10.1016/S0140-6736(12)60735-1 [DOI] [PubMed] [Google Scholar]
15.Cohen CI, Benyaminov R, Rahman M, Ngu D, Reinhardt M. Frailty: A Multidimensional Biopsychosocial Syndrome. Med Clin North Am. 2023;107(1):183–97. doi: 10.1016/j.mcna.2022.04.006 [DOI] [PubMed] [Google Scholar]
16.Casals C, Corral-Pérez J, Ávila-Cabeza-de-Vaca L, González-Mariscal A, Carrión-Velasco Y, Rodríguez-Martínez MC, et al. Exploring the interplay of frailty, physical function, physical activity, nutritional status, and their association with quality of life and depressive symptoms in older adults with the frailty phenotype. Int J Geriatr Psychiatry. 2024;39(3):e6078. doi: 10.1002/gps.6078 [DOI] [PubMed] [Google Scholar]
17.Bolton D. A revitalized biopsychosocial model: core theory, research paradigms, and clinical implications. Psychol Med. 2023;53(16):7504–11. doi: 10.1017/S0033291723002660 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Beyer A-K, Wolff JK, Warner LM, Schüz B, Wurm S. The role of physical activity in the relationship between self-perceptions of ageing and self-rated health in older adults. Psychol Health. 2015;30(6):671–85. doi: 10.1080/08870446.2015.1014370 [DOI] [PubMed] [Google Scholar]
19.Deshpande N, Metter EJ, Lauretani F, Bandinelli S, Guralnik J, Ferrucci L. Activity restriction induced by fear of falling and objective and subjective measures of physical function: a prospective cohort study. J Am Geriatr Soc. 2008;56(4):615–20. doi: 10.1111/j.1532-5415.2007.01639.x [DOI] [PMC free article] [PubMed] [Google Scholar]
20.John JM, Haug V, Thiel A. Physical Activity Behavior from a Transdisciplinary Biopsychosocial Perspective: a Scoping Review. Sports Med Open. 2020;6(1):49. doi: 10.1186/s40798-020-00279-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Komáromi L, Somogyi A, Szabo A. The Biopsychosocial Health Model Differentiates Long-Term Exercisers From Non-Exercisers: A Cross-Sectional Study. Health Sci Rep. 2025;8(2):e70476. doi: 10.1002/hsr2.70476 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Deng M-G, Liu F, Liang Y, Wang K, Nie J-Q, Liu J. Association between frailty and depression: A bidirectional Mendelian randomization study. Sci Adv. 2023;9(38):eadi3902. doi: 10.1126/sciadv.adi3902 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Liu R, Menhas R, Saqib ZA. Does physical activity influence health behavior, mental health, and psychological resilience under the moderating role of quality of life? Front Psychol. 2024;15:1349880. doi: 10.3389/fpsyg.2024.1349880 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Fan M, Lyu J, He P. Chinese guidelines for data processing and analysis concerning the International Physical Activity Questionnaire. Zhonghua Liu Xing Bing Xue Za Zhi. 2014;35(8):961–4. [PubMed] [Google Scholar]
25.Craig CL, Marshall AL, Sjöström M, Bauman AE, Booth ML, Ainsworth BE, et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc. 2003;35(8):1381–95. doi: 10.1249/01.MSS.0000078924.61453.FB [DOI] [PubMed] [Google Scholar]
26.Guralnik JM, Simonsick EM, Ferrucci L, Glynn RJ, Berkman LF, Blazer DG, et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol. 1994;49(2):M85-94. doi: 10.1093/geronj/49.2.m85 [DOI] [PubMed] [Google Scholar]
27.Katzman R, Zhang MY, Ouang Y-Q, Wang ZY, Liu WT, Yu E, et al. A Chinese version of the Mini-Mental State Examination; impact of illiteracy in a Shanghai dementia survey. J Clin Epidemiol. 1988;41(10):971–8. doi: 10.1016/0895-4356(88)90034-0 [DOI] [PubMed] [Google Scholar]
28.Wild B, Eckl A, Herzog W, Niehoff D, Lechner S, Maatouk I, et al. Assessing generalized anxiety disorder in elderly people using the GAD-7 and GAD-2 scales: results of a validation study. Am J Geriatr Psychiatry. 2014;22(10):1029–38. doi: 10.1016/j.jagp.2013.01.076 [DOI] [PubMed] [Google Scholar]
29.Cheung RYM. Patient Health Questionnaire-9 (PHQ-9). Handbook of Assessment in Mindfulness Research. Springer International Publishing; 2024. p. 1–12. doi: 10.1007/978-3-030-77644-2_63-2 [DOI] [Google Scholar]
30.Cauce AM, Felner RD, Primavera J. Social support in high-risk adolescents: structural components and adaptive impact. Am J Community Psychol. 1982;10(4):417–28. doi: 10.1007/BF00893980 [DOI] [PubMed] [Google Scholar]
31.Yan Z, Wang J, Dong Q, Zhu L, Lin W, Jiang X. XGBoost algorithm and logistic regression to predict the postoperative 5-year outcome in patients with glioma. Ann Transl Med. 2022;10(16):860. doi: 10.21037/atm-22-3384 [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Wang R, Wang L, Zhang J, He M, Xu J. XGBoost Machine Learning Algorism Performed Better Than Regression Models in Predicting Mortality of Moderate-to-Severe Traumatic Brain Injury. World Neurosurg. 2022;163:e617–22. doi: 10.1016/j.wneu.2022.04.044 [DOI] [PubMed] [Google Scholar]
33.Athanasiou M, Sfrintzeri K, Zarkogianni K, Thanopoulou AC, Nikita KS. An explainable XGBoost–based approach towards assessing the risk of cardiovascular disease in patients with Type 2 Diabetes Mellitus. In: 2020 IEEE 20th International Conference on Bioinformatics and Bioengineering (BIBE). 2020. p. 859–64. doi: 10.1109/bibe50027.2020.00146 [DOI] [Google Scholar]
34.Tanha J, Abdi Y, Samadi N, Razzaghi N, Asadpour M. Boosting methods for multi-class imbalanced data classification: an experimental review. J Big Data. 2020;7(1). doi: 10.1186/s40537-020-00349-y [DOI] [Google Scholar]
35.Lundberg SM, Lee SI. A unified approach to interpreting model predictions. Adv Neural Inf Process Syst. 2017;2017:4765–74. [Google Scholar]
36.Kunzmann U, Little TD, Smith J. Is age-related stability of subjective well-being a paradox? Cross-sectional and longitudinal evidence from the Berlin Aging Study. Psychol Aging. 2000;15(3):511–26. doi: 10.1037//0882-7974.15.3.511 [DOI] [PubMed] [Google Scholar]
37.Mayhew AJ, So HY, Ma J, Beauchamp MK, Griffith LE, Kuspinar A, et al. Normative values for grip strength, gait speed, timed up and go, single leg balance, and chair rise derived from the Canadian longitudinal study on ageing. Age Ageing. 2023;52(4):afad054. doi: 10.1093/ageing/afad054 [DOI] [PubMed] [Google Scholar]
38.Walston J, Buta B, Xue Q-L. Frailty Screening and Interventions: Considerations for Clinical Practice. Clin Geriatr Med. 2018;34(1):25–38. doi: 10.1016/j.cger.2017.09.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Hernandez HHC, Yang DZ, Tan CN, Kua J, Ismail NH, Lim WS. Short Physical Performance Battery Cutoff Points Using Clinical Outcomes for At-Risk Older Adults in Singapore: An Exploratory Study. Ann Geriatr Med Res. 2023;27(4):358–60. doi: 10.4235/agmr.23.0164 [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Lawton MP. The elderly in context: perspectives from environmental psychology and gerontology. Environment and Behavior. 1985;17(4):501–19. [Google Scholar]
41.Meijering L. Towards meaningful mobility: a research agenda for movement within and between places in later life. Ageing and Society. 2019;41(4):711–23. doi: 10.1017/s0144686x19001296 [DOI] [Google Scholar]
42.Taylor JK, Buchan IE, van der Veer SN. Assessing life-space mobility for a more holistic view on wellbeing in geriatric research and clinical practice. Aging Clin Exp Res. 2019;31(4):439–45. doi: 10.1007/s40520-018-0999-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Wu S, Wu W, Xia X, Zhou J. Characteristics of Physical Activities and Environmental Factor Preferences of Older Adults in Rural Resettlement Community in Ningbo, China. J Environ Public Health. 2022;2022:5414384. doi: 10.1155/2022/5414384 [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Yoshimura Y, Matsumoto A, Inoue T, Okamura M, Kuzuya M. Protein supplementation alone or combined with exercise for sarcopenia and physical frailty: A systematic review and meta-analysis of randomized controlled trials. Arch Gerontol Geriatr. 2025;131:105783. doi: 10.1016/j.archger.2025.105783 [DOI] [PubMed] [Google Scholar]
45.Yixiao C, Hui S, Quhong S, Xiaoxi Z, Jirong Y. A review of utility of wearable sensor technologies for older person frailty assessment. Exp Gerontol. 2025;200:112668. doi: 10.1016/j.exger.2024.112668 [DOI] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0335607.r001

Decision Letter 0

Hidetaka Hamasaki

23 Aug 2025

PONE-D-25-34585Factors contributing to differences in physical activity levels in (pre)frail older adults living in rural areas of ChinaPLOS ONE

Dear Dr. Zhang,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Oct 07 2025 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Hidetaka Hamasaki

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Your ethics statement should only appear in the Methods section of your manuscript. If your ethics statement is written in any section besides the Methods, please move it to the Methods section and delete it from any other section. Please ensure that your ethics statement is included in your manuscript, as the ethics statement entered into the online submission form will not be published alongside your manuscript.

3. In the online submission form, you indicated that data cannot be shared publicly because of ethical consideration. Data are available from the corresponding author Institutional Data Access / Ethics Committee for researchers who meet the criteria for access to confidential data.

All PLOS journals now require all data underlying the findings described in their manuscript to be freely available to other researchers, either 1. In a public repository, 2. Within the manuscript itself, or 3. Uploaded as supplementary information.

This policy applies to all data except where public deposition would breach compliance with the protocol approved by your research ethics board. If your data cannot be made publicly available for ethical or legal reasons (e.g., public availability would compromise patient privacy), please explain your reasons on resubmission and your exemption request will be escalated for approval.

4. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

5. Please note that funding information should not appear in any section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form. Please remove any funding-related text from the manuscript.

6. If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: No

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Introduction:

1.Literature gap justification: The introduction mentions that "it is not yet clear how these factors are related to physical activity levels in pre(frail) older adults especially those living in rural areas." However, could you provide more specific evidence about what aspects of this relationship remain unexplored, given the existing literature on frailty and physical activity?

2.Hypothesis development: The hypothesis states that "self-perceived health will have a major impact on physical activity levels compared to other health-related factors." What theoretical framework or preliminary evidence supports prioritizing self-perceived health over other established predictors like social support or physical performance?

3.Geographic specificity: The introduction emphasizes rural China, but how do the unique characteristics of Northeast China (climate, culture, economic conditions) potentially influence the generalizability of findings to other rural contexts?

Methods

4.Frailty classification consistency: The inclusion criteria require "frailty score >0 according to the Fried Phenotype," but later analysis combines pre-frail and frail participants. How might this heterogeneity within the sample affect the interpretation of physical activity predictors, and should these groups be analyzed separately?

5. IPAQ-SF validity concerns: Given that you acknowledge in the discussion that IPAQ-SF tends to overestimate activity levels with correlations often below acceptable standards, why was this instrument chosen over more objective measures, and how might this limitation affect your primary findings?

6.XGBoost model justification: While you mention XGBoost can handle imbalanced data and non-linear relationships, what specific advantages does this approach offer over traditional logistic regression for this particular research question, especially given the relatively small sample size (n=284)?

Results

7.Class imbalance impact: Despite using class weighting, your model shows 90% specificity but only 63% sensitivity. How does this imbalance affect the practical utility of the model, particularly for identifying individuals who might benefit from physical activity interventions?

8.Feature importance interpretation: The SHAP analysis shows SSRS (social support) as highly influential, but Figure 4 shows considerable overlap between groups. How do you explain this apparent contradiction, and what does this suggest about the model's ability to capture the true relationship?

9. Physical performance paradox: Your results show that even the high physical activity group performed below the 5th percentile for age-matched peers on physical tests. How do you reconcile high self-reported physical activity levels with poor objective physical performance measures?

Discussion

10. Causality concerns: The discussion frequently implies causal relationships (e.g., "self-reported health, social support, and physical activity engagement"), but this is a cross-sectional study. How do you address the potential for reverse causality, where poor physical function might lead to negative health perceptions rather than vice versa?

11.Intervention implications: You suggest "community-based social support programs and structured mobility-enhancing exercises," but given that most participants already met high physical activity levels (despite poor physical performance), what specific interventions would be most appropriate for this population?

12.Methodological limitations: While you acknowledge IPAQ-SF limitations in the discussion, this appears to be a fundamental flaw that may compromise the study's validity. How significantly might measurement error affect your conclusions, and should this limitation be addressed more prominently earlier in the manuscript?

Reviewer #2: This manuscript describes the study of factors contributing to differences in physical activity levels in prefrail older adults in rural areas This is an important topic since frailty prevention in older adults should be a priority in all populations.

General comments:

- Authors should follow financial disclosure guidelines.

- Language revision is required.

- Authors declared all data are available without restrictions, yet they later stated that data cannot be shared publicly. Non-identifiable data can be provided without restrictions, and the statements need to be congruent.

Abstract:

- Abstract sections do not usually include a discussion, instead; they include a conclusion. Please follow Plos One guidelines.

Introduction:

- I am not sure how the Chinese government classifies age groups. Is the definition of older adults starts at the age 60 or 65? The authors included patients who were 60 or above. Yet, in the introduction, the classification states 65 years or above. Please explain why you chose to use the age 60 or more instead of official classification.

- Aim: Authors intended to measure the influence of different health domains on physical activity levels. Since this was a cross-sectional study, causality cannot be proven. All we can confirm here is the presence of an association. I would assume that higher physical activity can positively influence other health domains. This is more logically proven via studies on the positive influences of higher physical activity on many health-related aspects and on diseases.

Methods:

- Including all those with Fried’s criteria scores more than 0 means including patients with frailty as well.

- Please provide the approval to use MMSE as it is copy righted.

- Authors included anxiety, depression and social support under perceived health item in the results. Psychiatric problems are not the only contributors to health and not including physical problems as well can gravely affect the results. Please include other aspects of health in your analysis.

Results:

- Where are the supporting tables? I was not able to find them.

- In table 1: authors did not put the significance of all items. Only significant variables had their p values in the table. Please add all the data in the table.

Discussion:

- Please cite the authors by name when it is part of the sentence structure (Page 25 line 324).

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2025 Nov 4;20(11):e0335607. doi: 10.1371/journal.pone.0335607.r002

Author response to Decision Letter 1

25 Sep 2025

Please refer to the "response to the reviwer.docx", find the point-to-point answer.

Attachment

Submitted filename: Response to Reviewers.docx

pone.0335607.s004.docx^{(72.4KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0335607.r003

Decision Letter 1

Hidetaka Hamasaki

14 Oct 2025

Factors contributing to differences in physical activity levels in (pre)frail older adults living in rural areas of China

PONE-D-25-34585R1

Dear Dr. Zhang,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at Editorial Manager® and clicking the ‘Update My Information' link at the top of the page. For questions related to billing, please contact billing support.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Hidetaka Hamasaki

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Thank you for submitting the revised manuscript.

Since there has been no response from the two reviewers, I have checked your responses to the reviewers as the Academic Editor on their behalf. It appears that you have provided careful and detailed replies to the reviewers’ comments and suggestions. Appropriate revisions have also been made to the main text and tables, and I believe the manuscript meets the publication standard.

Reviewers' comments:

PLoS One. doi: 10.1371/journal.pone.0335607.r004

Acceptance letter

Hidetaka Hamasaki

PONE-D-25-34585R1

PLOS ONE

Dear Dr. Zhang,

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now being handed over to our production team.

At this stage, our production department will prepare your paper for publication. This includes ensuring the following:

* All references, tables, and figures are properly cited

* All relevant supporting information is included in the manuscript submission,

* There are no issues that prevent the paper from being properly typeset

You will receive further instructions from the production team, including instructions on how to review your proof when it is ready. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few days to review your paper and let you know the next and final steps.

Lastly, if your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

You will receive an invoice from PLOS for your publication fee after your manuscript has reached the completed accept phase. If you receive an email requesting payment before acceptance or for any other service, this may be a phishing scheme. Learn how to identify phishing emails and protect your accounts at https://explore.plos.org/phishing.

If we can help with anything else, please email us at customercare@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Hidetaka Hamasaki

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 File. Supplementary material-appendixes.

(DOCX)

pone.0335607.s001.docx^{(26.4KB, docx)}

S2 File. Data XGBOOST.

(CSV)

pone.0335607.s002.csv^{(22.5KB, csv)}

Attachment

Submitted filename: Response to Reviewers.docx

pone.0335607.s004.docx^{(72.4KB, docx)}

Data Availability Statement

All relevant data are within the paper and its Supporting Information files.

[pone.0335607.ref001] 1.Feng Z, Glinskaya E, Chen H, Gong S, Qiu Y, Xu J, et al. Long-term care system for older adults in China: policy landscape, challenges, and future prospects. Lancet. 2020;396(10259):1362–72. doi: 10.1016/S0140-6736(20)32136-X [DOI] [PubMed] [Google Scholar]

[pone.0335607.ref002] 2.Deng SY, Zhao IY, Ho M-H, Saravanakumar P, Molassiotis A, Montayre J. Rural-urban disparities in healthy ageing: Evidence from a national study in China. Collegian. 2022;29(6):843–9. doi: 10.1016/j.colegn.2022.06.008 [DOI] [Google Scholar]

[pone.0335607.ref003] 3.Li H, Zeng Y, Gan L, Tuersun Y, Yang J, Liu J, et al. Urban-rural disparities in the healthy ageing trajectory in China: a population-based study. BMC Public Health. 2022;22(1):1406. doi: 10.1186/s12889-022-13757-x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref004] 4.Xu R, Li Q, Guo F, Zhao M, Zhang L. Prevalence and risk factors of frailty among people in rural areas: a systematic review and meta-analysis. BMJ Open. 2021;11(4):e043494. doi: 10.1136/bmjopen-2020-043494 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref005] 5.Zeng X-Z, Meng L-B, Li Y-Y, Jia N, Shi J, Zhang C, et al. Prevalence and factors associated with frailty and pre-frailty in the older adults in China: a national cross-sectional study. Front Public Health. 2023;11:1110648. doi: 10.3389/fpubh.2023.1110648 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref006] 6.Zhou Q, Li Y, Gao Q, Yuan H, Sun L, Xi H, et al. Prevalence of Frailty Among Chinese Community-Dwelling Older Adults: A Systematic Review and Meta-Analysis. Int J Public Health. 2023;68:1605964. doi: 10.3389/ijph.2023.1605964 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref007] 7.Dent E, Martin FC, Bergman H, Woo J, Romero-Ortuno R, Walston JD. Management of frailty: opportunities, challenges, and future directions. Lancet. 2019;394(10206):1376–86. doi: 10.1016/S0140-6736(19)31785-4 [DOI] [PubMed] [Google Scholar]

[pone.0335607.ref008] 8.Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci. 2001;56(3):M146-56. doi: 10.1093/gerona/56.3.m146 [DOI] [PubMed] [Google Scholar]

[pone.0335607.ref009] 9.Muntner P, Gu D, Wildman RP, Chen J, Qan W, Whelton PK, et al. Prevalence of physical activity among Chinese adults: results from the International Collaborative Study of Cardiovascular Disease in Asia. Am J Public Health. 2005;95(9):1631–6. doi: 10.2105/AJPH.2004.044743 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref010] 10.Jiménez-Ten Hoevel C, Queral J, Besora-Moreno M, Tarro L, Sospedra-Senar A, Tomàs-Piqueras J, et al. Health status, lifestyle and quality of life in older adults of rural and urban areas of Tarragona province (Catalonia, Spain): a cross-sectional descriptive study. BMC Public Health. 2025;25(1):895. doi: 10.1186/s12889-025-22045-3 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref011] 11.Furtado GE, Vaz C, Bovolini A, Marques E, Serra N, Costa-Brito AR, et al. The Impact of Physical Activity Levels and Functional Fitness Status on the Quality of Life Perceived by Older Adults Living in Rural and Urban Areas: The Portuguese Inland Case. Healthcare (Basel). 2022;10(7):1266. doi: 10.3390/healthcare10071266 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref012] 12.Wang Y, Wang X, Zhu X, Sun Y, Han B, Chen T. Physical activity, physical frailty and depressive symptoms among Chinese male and female older adults: do different indicators make a difference?. Front Public Health. 2023;11:1280952. doi: 10.3389/fpubh.2023.1280952 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref013] 13.Steinhoff P, Reiner A. Physical activity and functional social support in community-dwelling older adults: a scoping review. BMC Public Health. 2024;24(1):1355. doi: 10.1186/s12889-024-18863-6 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref014] 14.Bauman AE, Reis RS, Sallis JF, Wells JC, Loos RJF, Martin BW, et al. Correlates of physical activity: why are some people physically active and others not?. Lancet. 2012;380(9838):258–71. doi: 10.1016/S0140-6736(12)60735-1 [DOI] [PubMed] [Google Scholar]

[pone.0335607.ref015] 15.Cohen CI, Benyaminov R, Rahman M, Ngu D, Reinhardt M. Frailty: A Multidimensional Biopsychosocial Syndrome. Med Clin North Am. 2023;107(1):183–97. doi: 10.1016/j.mcna.2022.04.006 [DOI] [PubMed] [Google Scholar]

[pone.0335607.ref016] 16.Casals C, Corral-Pérez J, Ávila-Cabeza-de-Vaca L, González-Mariscal A, Carrión-Velasco Y, Rodríguez-Martínez MC, et al. Exploring the interplay of frailty, physical function, physical activity, nutritional status, and their association with quality of life and depressive symptoms in older adults with the frailty phenotype. Int J Geriatr Psychiatry. 2024;39(3):e6078. doi: 10.1002/gps.6078 [DOI] [PubMed] [Google Scholar]

[pone.0335607.ref017] 17.Bolton D. A revitalized biopsychosocial model: core theory, research paradigms, and clinical implications. Psychol Med. 2023;53(16):7504–11. doi: 10.1017/S0033291723002660 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref018] 18.Beyer A-K, Wolff JK, Warner LM, Schüz B, Wurm S. The role of physical activity in the relationship between self-perceptions of ageing and self-rated health in older adults. Psychol Health. 2015;30(6):671–85. doi: 10.1080/08870446.2015.1014370 [DOI] [PubMed] [Google Scholar]

[pone.0335607.ref019] 19.Deshpande N, Metter EJ, Lauretani F, Bandinelli S, Guralnik J, Ferrucci L. Activity restriction induced by fear of falling and objective and subjective measures of physical function: a prospective cohort study. J Am Geriatr Soc. 2008;56(4):615–20. doi: 10.1111/j.1532-5415.2007.01639.x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref020] 20.John JM, Haug V, Thiel A. Physical Activity Behavior from a Transdisciplinary Biopsychosocial Perspective: a Scoping Review. Sports Med Open. 2020;6(1):49. doi: 10.1186/s40798-020-00279-2 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref021] 21.Komáromi L, Somogyi A, Szabo A. The Biopsychosocial Health Model Differentiates Long-Term Exercisers From Non-Exercisers: A Cross-Sectional Study. Health Sci Rep. 2025;8(2):e70476. doi: 10.1002/hsr2.70476 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref022] 22.Deng M-G, Liu F, Liang Y, Wang K, Nie J-Q, Liu J. Association between frailty and depression: A bidirectional Mendelian randomization study. Sci Adv. 2023;9(38):eadi3902. doi: 10.1126/sciadv.adi3902 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref023] 23.Liu R, Menhas R, Saqib ZA. Does physical activity influence health behavior, mental health, and psychological resilience under the moderating role of quality of life? Front Psychol. 2024;15:1349880. doi: 10.3389/fpsyg.2024.1349880 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref024] 24.Fan M, Lyu J, He P. Chinese guidelines for data processing and analysis concerning the International Physical Activity Questionnaire. Zhonghua Liu Xing Bing Xue Za Zhi. 2014;35(8):961–4. [PubMed] [Google Scholar]

[pone.0335607.ref025] 25.Craig CL, Marshall AL, Sjöström M, Bauman AE, Booth ML, Ainsworth BE, et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc. 2003;35(8):1381–95. doi: 10.1249/01.MSS.0000078924.61453.FB [DOI] [PubMed] [Google Scholar]

[pone.0335607.ref026] 26.Guralnik JM, Simonsick EM, Ferrucci L, Glynn RJ, Berkman LF, Blazer DG, et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol. 1994;49(2):M85-94. doi: 10.1093/geronj/49.2.m85 [DOI] [PubMed] [Google Scholar]

[pone.0335607.ref027] 27.Katzman R, Zhang MY, Ouang Y-Q, Wang ZY, Liu WT, Yu E, et al. A Chinese version of the Mini-Mental State Examination; impact of illiteracy in a Shanghai dementia survey. J Clin Epidemiol. 1988;41(10):971–8. doi: 10.1016/0895-4356(88)90034-0 [DOI] [PubMed] [Google Scholar]

[pone.0335607.ref028] 28.Wild B, Eckl A, Herzog W, Niehoff D, Lechner S, Maatouk I, et al. Assessing generalized anxiety disorder in elderly people using the GAD-7 and GAD-2 scales: results of a validation study. Am J Geriatr Psychiatry. 2014;22(10):1029–38. doi: 10.1016/j.jagp.2013.01.076 [DOI] [PubMed] [Google Scholar]

[pone.0335607.ref029] 29.Cheung RYM. Patient Health Questionnaire-9 (PHQ-9). Handbook of Assessment in Mindfulness Research. Springer International Publishing; 2024. p. 1–12. doi: 10.1007/978-3-030-77644-2_63-2 [DOI] [Google Scholar]

[pone.0335607.ref030] 30.Cauce AM, Felner RD, Primavera J. Social support in high-risk adolescents: structural components and adaptive impact. Am J Community Psychol. 1982;10(4):417–28. doi: 10.1007/BF00893980 [DOI] [PubMed] [Google Scholar]

[pone.0335607.ref031] 31.Yan Z, Wang J, Dong Q, Zhu L, Lin W, Jiang X. XGBoost algorithm and logistic regression to predict the postoperative 5-year outcome in patients with glioma. Ann Transl Med. 2022;10(16):860. doi: 10.21037/atm-22-3384 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref032] 32.Wang R, Wang L, Zhang J, He M, Xu J. XGBoost Machine Learning Algorism Performed Better Than Regression Models in Predicting Mortality of Moderate-to-Severe Traumatic Brain Injury. World Neurosurg. 2022;163:e617–22. doi: 10.1016/j.wneu.2022.04.044 [DOI] [PubMed] [Google Scholar]

[pone.0335607.ref033] 33.Athanasiou M, Sfrintzeri K, Zarkogianni K, Thanopoulou AC, Nikita KS. An explainable XGBoost–based approach towards assessing the risk of cardiovascular disease in patients with Type 2 Diabetes Mellitus. In: 2020 IEEE 20th International Conference on Bioinformatics and Bioengineering (BIBE). 2020. p. 859–64. doi: 10.1109/bibe50027.2020.00146 [DOI] [Google Scholar]

[pone.0335607.ref034] 34.Tanha J, Abdi Y, Samadi N, Razzaghi N, Asadpour M. Boosting methods for multi-class imbalanced data classification: an experimental review. J Big Data. 2020;7(1). doi: 10.1186/s40537-020-00349-y [DOI] [Google Scholar]

[pone.0335607.ref035] 35.Lundberg SM, Lee SI. A unified approach to interpreting model predictions. Adv Neural Inf Process Syst. 2017;2017:4765–74. [Google Scholar]

[pone.0335607.ref036] 36.Kunzmann U, Little TD, Smith J. Is age-related stability of subjective well-being a paradox? Cross-sectional and longitudinal evidence from the Berlin Aging Study. Psychol Aging. 2000;15(3):511–26. doi: 10.1037//0882-7974.15.3.511 [DOI] [PubMed] [Google Scholar]

[pone.0335607.ref037] 37.Mayhew AJ, So HY, Ma J, Beauchamp MK, Griffith LE, Kuspinar A, et al. Normative values for grip strength, gait speed, timed up and go, single leg balance, and chair rise derived from the Canadian longitudinal study on ageing. Age Ageing. 2023;52(4):afad054. doi: 10.1093/ageing/afad054 [DOI] [PubMed] [Google Scholar]

[pone.0335607.ref038] 38.Walston J, Buta B, Xue Q-L. Frailty Screening and Interventions: Considerations for Clinical Practice. Clin Geriatr Med. 2018;34(1):25–38. doi: 10.1016/j.cger.2017.09.004 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref039] 39.Hernandez HHC, Yang DZ, Tan CN, Kua J, Ismail NH, Lim WS. Short Physical Performance Battery Cutoff Points Using Clinical Outcomes for At-Risk Older Adults in Singapore: An Exploratory Study. Ann Geriatr Med Res. 2023;27(4):358–60. doi: 10.4235/agmr.23.0164 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref040] 40.Lawton MP. The elderly in context: perspectives from environmental psychology and gerontology. Environment and Behavior. 1985;17(4):501–19. [Google Scholar]

[pone.0335607.ref041] 41.Meijering L. Towards meaningful mobility: a research agenda for movement within and between places in later life. Ageing and Society. 2019;41(4):711–23. doi: 10.1017/s0144686x19001296 [DOI] [Google Scholar]

[pone.0335607.ref042] 42.Taylor JK, Buchan IE, van der Veer SN. Assessing life-space mobility for a more holistic view on wellbeing in geriatric research and clinical practice. Aging Clin Exp Res. 2019;31(4):439–45. doi: 10.1007/s40520-018-0999-5 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref043] 43.Wu S, Wu W, Xia X, Zhou J. Characteristics of Physical Activities and Environmental Factor Preferences of Older Adults in Rural Resettlement Community in Ningbo, China. J Environ Public Health. 2022;2022:5414384. doi: 10.1155/2022/5414384 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0335607.ref044] 44.Yoshimura Y, Matsumoto A, Inoue T, Okamura M, Kuzuya M. Protein supplementation alone or combined with exercise for sarcopenia and physical frailty: A systematic review and meta-analysis of randomized controlled trials. Arch Gerontol Geriatr. 2025;131:105783. doi: 10.1016/j.archger.2025.105783 [DOI] [PubMed] [Google Scholar]

[pone.0335607.ref045] 45.Yixiao C, Hui S, Quhong S, Xiaoxi Z, Jirong Y. A review of utility of wearable sensor technologies for older person frailty assessment. Exp Gerontol. 2025;200:112668. doi: 10.1016/j.exger.2024.112668 [DOI] [PubMed] [Google Scholar]

PERMALINK

Factors contributing to differences in physical activity levels in (pre)frail older adults living in rural areas of China

Xin Zhang

Xiaoping Zheng

Hans Hobbelen

Barbara van Munster

Qian Tong

Tianzhuo Yu

Feng Li

Claudine JC Lamoth

Roles

Abstract

Introduction

Methods

Results

Conclusion

1. Introduction

2. Methods

2.1 Study design

2.2 Population

2.3 Assessment instruments

Fig 1. The data processing and analysis pipeline.

2.4 Data analysis

Fig 2. Difference of overall MET values and variables derived from physical performance tests between low-moderate and high PA levels group.

2.4.1 Architecture and pipeline of XGBoost.

2.4.2 Model performance assessment.

2.4.3 Feature importance.

3. Results

3.1 General characteristics

3.2 Difference of observed variables between low-moderate and high PA group

Table 1. Demographics and outcome variables of the two groups.

Fig 3. Cognitive test and self-perceived health and social functioning presented for the low-moderate and high PA levels group.

3.3 Performance of XGBoost model

Fig 4. The SHAP (Shapley Additive exPlanations) values of observed variables.

4. Discussion

5. Conclusions

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Hidetaka Hamasaki

Roles

Author response to Decision Letter 1

Decision Letter 1

Hidetaka Hamasaki

Roles

Acceptance letter

Hidetaka Hamasaki

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases