Comparison of logistic regression and machine learning methods for predicting depression risks among disabled elderly individuals: results from the China Health and Retirement Longitudinal Study

Comparison of logistic regression and machine learning methods for predicting depression risks among disabled elderly individuals: results from the China Health and Retirement Longitudinal Study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Comparison of logistic regression and machine learning methods for predicting depression risks among disabled elderly individuals: results from the China Health and Retirement Longitudinal Study Shanshan Hong, Bingqian Lu, Shaobing Wang, Yan Jiang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3874875/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 14 Feb, 2025 Read the published version in BMC Psychiatry → Version 1 posted 12 You are reading this latest preprint version Abstract Background Given the accelerated aging population in China, the number of disabled elderly individuals is increasing, depression has been a common mental disorder among older adults. This study aims to establish an effective model for predicting depression risks among disabled elderly individuals. Methods The data for this study was obtained from the 2018 China Health and Retirement Longitudinal Study (CHARLS). In this study, disability was defined as a functional impairment in at least one activity of daily living (ADL) or instrumental activity of daily living (IADL). Depressive symptoms were assessed by using the 10-item Center for Epidemiologic Studies Depression Scale (CES-D10). We employed SPSS 27.0 to select independent risk factor variables associated with depression among disabled elderly individuals. Subsequently, a predictive model for depression in this population was constructed using R 4.3.0. The model's discrimination, calibration, and clinical net benefits were assessed using receiver operating characteristic (ROC) curves, calibration plots, and decision curves. Results In this study, a total of 3,107 elderly individuals aged ≥ 60 years with disabilities were included. Poor self-rated health, pain, absence of caregivers, cognitive impairment, and shorter sleep duration were identified as independent risk factors for depression in disabled elderly individuals. The XGBoost model demonstrated better predictive performance in the training set, while the logistic regression model showed better predictive performance in the validation set, with AUC of 0.76 and 0.73, respectively. The calibration curve and Brier score (Brier: 0.20) indicated a good model fit. Moreover, decision curve analysis confirmed the clinical utility of the model. Conclusions The predictive model exhibits outstanding predictive efficacy, greatly assisting healthcare professionals and family members in evaluating depression risks among disabled elderly individuals. Consequently, it enables the early identification of elderly individuals at high risks for depression. Disability Depression Predictive Model Elderly individuals Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Background Research indicates that the population of disabled elderly individuals in China has exceeded 40 million, making up over 16% of the elderly population. Projections anticipate that the number will reach 65 million by 2030 [ 1 ]. Depression is a prevalent mental health disorder among the elderly population, impacting around 7% of them worldwide [ 2 , 3 ]. Both disability and depression have emerged as notable social concerns. Disability encompasses physical, psychological, and social aspects and pertains to the impaired capacity of older individuals to engage in fundamental activities of daily living autonomously. This decline in functional abilities hampers their adaptation to environmental changes. Depression is a common psychological issue characterized by a chronic trajectory and a high likelihood of recurrence [ 4 ]. It is associated with various adverse outcomes, such as reduced quality of life, amplified healthcare burden, and heightened incidence and mortality rates [ 5 ]. Both disability and depression impose substantial burdens on individuals as well as society. Late-life depression is closely associated with disability, particularly when it hinders self-care and social participation [ 6 ]. Research indicates that the likelihood of depression in the population with physical disabilities is at least three times higher compared to normal individuals [ 7 ]. Moreover, there is a potential direct causal link between disability among older adults and the occurrence of depression. There could exist a direct causal relationship between disability and depression in the older adult population [ 8 , 9 ]. Previous research has indicated that several significant factors are associated with the onset and persistence of depression in older adults, including being female, having a low educational level, experiencing spousal loss, cognitive decline, physical illness, and functional impairment [ 10 ]. Currently, the relationship between disability in elderly individuals and depression is still the subject of ongoing investigations, and interventions for depression among disabled older adults are considered essential for healthcare services. Clinical risk prediction models have become widely prevalent in various medical fields in recent years. These models aim to utilize individual-level information to predict clinically relevant outcomes [ 11 ]. By employing clinical prediction models, the efficiency of healthcare processes and self-management can be enhanced while reducing the workforce costs associated with healthcare [ 12 ]. In certain studies, relevant biochemical indicators such as HDL-C, fasting blood glucose, triglycerides, and other metabolic markers are integrated into prediction models [ 13 ]. However, the process of collecting data for these indicators is invasive or costly, necessitating well-trained experts and controlled experimental environments, often rendering them impractical in community settings. While the inclusion of biochemical indicators may enhance the predictive performance of risk models, it may also diminish their practicality [ 14 ]. Primary healthcare demands more straightforward tools [ 15 ]. In primary healthcare settings, risk prediction models based on readily accessible data concerning risk factors associated with depression in disabled elderly individuals prove to be more suitable and feasible [ 16 ]. We constructed a clinical prediction model for depression among disabled elderly individuals by analyzing data from the 2018 China Health and Retirement Longitudinal Study (CHARLS) to identify the relevant factors associated with depression in this population. By utilizing commonly available predictive factors in the community environment, we developed a practical prediction system for assessing the risks of depression among disabled elderly individuals. This system aims to assist healthcare professionals in measuring the probability of depression in this population and facilitate to identify the risk of depression at an early stage among these individuals. Methods Study design and participants The data was derived from the China Health and Retirement Longitudinal Study (CHARLS). CHARLS is conducted every two years (2011, 2013, 2015, 2018), with each wave adding new participants. The CHARLS participants were sampled using a multistage probability sampling strategy and probability proportionate to the size sampling method. It covered 150 counties of 28 provinces, municipal cities, and autonomous regions of China [ 17 ]. The CHARLS was approved by the Ethical Review Committee at Peking University, and all participants signed informed consent before participation. We employed the baseline data from 2018 to develop a practical risk prediction model for depressive symptoms among disabled older adults. In the longitudinal CHARLS cohort, disability is classified into two dimensions: Activities of Daily Living (ADL) and Instrumental Activities of Daily Living (IADL). To assess disability, this study utilized the PSMS scale and Lawton scale, each comprising six sections[ 18 ]. In the present study, participants who acknowledge difficulties by responding with “yes, I have difficulty and need assistance”, or express an inability with “I cannot do it” in the context of any given project, are regarded as functionally impaired[ 19 , 20 ]. Participants were considered disabled if any section was classified as such. The 2018 baseline survey included a total of 19,817 participants. After excluding individuals with missing or abnormal data, the final analytical sample consisted of 3,107 disabled older adults aged 60 years and above. Figure 1 illustrates the flowchart for selecting and following up on all eligible study subjects. Measures Depressive symptoms In the CHARLS cohort, depression is measured by using the 10-item Center for Epidemiologic Studies Depression Scale (CESD-10) administered through the survey [ 21 ]. This scale consists of 10 components: "feeling down," "feeling afraid," "feeling lonely," "having poor sleep," "having difficulty in completing tasks," "being irritable over minor matters," "having trouble concentrating" and "feeling unable to continue with life." The items "I have hope for the future" and "I feel happy" in this scale are scored inversely. Each question offers four response options: Rarely or none of the time (< 1 day), Some or a little of the time (1–2 days), Occasionally or a moderate amount of the time (3–4 days), and most or all of the time (5–7 days).The overall scale score ranges from 0 to 30, with a threshold of 10 points indicating a tendency towards depression. Variables Considering the accuracy and practicality of the risk prediction model, we have incorporated variables related to depression among disabled older adults based on a review of existing evidence [ 6 , 9 , 22 , 23 ]. Relevant information was collected through structured questionnaires, including sociodemographic factors (age, gender, place of residence, educational level, marital status, etc), health behaviors (smoking, drinking history, sleep, etc.), cognitive status and social support. Statistical analysis Data cleaning, preprocessing and merging were conducted using the STATA 17.0 software. The analysis of influencing factors and determining important predictive factors were performed on the sample using SPSS 27.0. The chi-square test for categorical variables was employed to compare the influencing factors of depression among disabled older adults and select variables with statistical significance. Important predictive factors related to depression among disabled older adults were selected based on expert opinions and through univariate and multivariate logistic regression analysis using the forward stepwise selection method. The statistical significance level was set at 0.05 for two-tailed tests. Model construction and evaluation A depression risk prediction model was constructed for elderly individuals with disabilities through the utilization of logistic regression and four machine learning algorithms: Support Vector Machine (SVM), Random Forest (RF), XGBoost, and Decision Tree. During the modeling process, conducting 10-fold cross-validation ensures the performance and stability of the model. The model's development was facilitated using R version 4.3.0. Create the receiver operating characteristic (ROC) curve to evaluate the predictive capability of the prediction model, where an area under the curve (AUC) value of ≥ 0.70 is considered appropriate [ 24 ]. The calibration curve visualizes the consistency assessment between the average predicted risk occurrence rate and the actual observed event occurrence rate. If the solid line (representing the performance of the predictive model) is closer to the diagonal dashed line (representing the perfect prediction of an ideal model), it indicates better consistency [ 25 , 26 ]. A Brier score of less than 0.25 indicates that the overall performance evaluation of discrimination and calibration is acceptable [ 27 ]. The decision curve evaluates the clinical utility by assessing the net benefit at different threshold probabilities [ 28 ]. Model interpretation SHAP, an acronym for Shapley Additive exPlanations, was introduced by Lundberg and Lee in 2017[ 29 ]. It constitutes a framework employed for explicating the predictions made by machine learning models to elucidate each feature's contribution[ 30 ]. Shapley values, originating from cooperative game theory, serve as an equitable approach for allocating gains in cooperative games. About participants in cooperative games, Shapley values contemplate the marginal contributions of each participant, thereby ensuring a judicious distribution of benefits[ 31 ]. In machine learning, features can be construed as participants in a cooperative game, with the model's output representing the dividends of the game[ 32 ]. Results Participant characteristics A total of 3,107 disabled elderly individuals were included in this study. Among them, 1,774 (57.1%) had depression, making up 57.1% of the total sample of disabled elderly individuals. Within the training set, 56.66% (1233/2176) of individuals exhibited depression, whereas in the validation set, the proportion was 58.11% (541/931). Characteristics of the study population are given in Table 1 . Table 1 Characteristics of the participants at baseline Variable Total Non-depressive depression x 2 p n = 3107 n = 1333 n = 1774 Gender (%) male female 1278(41.13) 1829(58.87) 623(46.74) 710(53.26) 655(36.9) 1119(63.08) 30.28 < 0.001 Age, years 60–69 70–79 ≥ 80 1726(55.55) 1101(35.44) 280(9.01) 711(53.34) 476(35.71) 146(10.95) 1015(57.22) 625(35.23) 134(7.55) 11.867 0.003 Residence village non-rural 2460(79.18) 647(20.82) 1013(75.99) 320(24.01) 1447(81.57) 327(18.43) 14.34 < 0.001 Education level illiteracy elementary school level junior high school level high school and above 1143(36.79) 1429(46.00) 405(13.04) 130(4.18) 462(34.66) 597(44.79) 198(14.85) 76(5.70) 681(38.39) 832(46.90) 207(11.67) 54(3.04) 22.386 < 0.001 Health self-assessment good common poor 340(10.94) 1238(39.85) 1529(49.21) 234(17.55) 638(47.86) 461(34.58) 106(5.98) 600(33.82) 1068(60.20) 232.416 < 0.001 pains none occasionally often 668(21.50) 1339(43.10) 1100(35.40) 424(31.81) 619(46.44) 290(21.76) 244(13.75) 720(40.59) 810(45.66) 244.266 < 0.001 Sleep at night 8h 1454(46.80) 1328(42.74) 325(10.46) 462(34.66) 707(53.04) 164(12.30) 992(55.92) 621(35.01) 161(9.08) 138.99 < 0.001 Social activities no yes 1689(54.36) 1418(45.64) 680(51.01) 653(48.99) 1009(56.88) 765(43.12) 10.55 < 0.001 Caregiver yes no 2094(67.40) 1013(32.60) 1004(75.32) 329(24.68) 1090(61.44) 684(38.56) 66.69 < 0.001 Work no problem short time unable 965(31.06) 1250(40.23) 892(28.71) 541(40.59) 477(35.78) 315(23.63) 424(23.90) 773(43.57) 577(32.53) 100.67 < 0.001 Cognitive impairment no yes 1357(43.68) 1750(56.32) 644(48.31) 689(51.69) 713(40.19) 1061(59.81) 20.402 5000 2615(84.16) 492(15.84) 1068(80.12) 265(19.88) 1547(87.20) 227(12.80) 28.658 < 0.001 Variable selection results Table 1 presents the influencing factors of depression in disabled elderly. Univariate logistic regression and multivariate logistic regression were employed to identify independent risk factors for depression among disabled elderly, resulting in the selection of five significant predictive factors for constructing a depression prediction model in disabled elderly, as shown in Tables 2 and 3 . " poor self-rated health", "pain", " shorter sleep duration at night", "lack of caregivers" and "cognitive impairment" emerged as independent risk factors in the logistic regression analysis. Table 2 Results of univariate logistic regression analysis B S.E Wald df Sig. Exp(B) 95%C.I.for EXP(B) Gender -0.510 0.088 33.618 1 =80 0.444 0.154 8.256 1 0.004 1.558 1.151–2.109 70–79 vs. >=80 0.350 0.161 4.753 1 0.029 1.419 1.036–1.994 Residence 0.413 0.106 15.294 1 < 0.001 1.512 1.229–1.859 Education level 25.552 3 < 0.001 illiteracy vs. high school and above 0. 978 0.229 18.244 1 < 0.001 2.660 1.698–4.166 elementary school level vs. high school and above 0.910 0.227 16.111 1 < 0.001 2.483 1.593–3.872 junior high school level vs. high school and above 0.553 0.248 4.971 1 0.026 1.738 1.069–2.825 Health self-assessment 162.176 2 < 0.001 good vs. poor -1.534 0.154 98.787 1 < 0.001 0.216 0.159–0.292 common vs. poor -1.004 0.095 110.951 1 < 0.001 0.367 0.304–0.442 Pain 178.705 2 < 0.001 none vs. often -1.669 0.127 172.755 1 < 0.001 0.188 0.147–0.242 Occasionally vs. often -0.942 0.106 79.041 1 < 0.001 0.390 0.317–0.480 Sleep at night 86.648 2 < 0.001 8h 0.726 0.148 24.217 1 8h -0.126 0.147 0.733 1 0.392 0.882 0.662–1.176 Social activities -0.249 0.087 8.219 1 0.004 0.779 0.657–0.924 Caregiver -0.592 0.095 38.790 1 < 0.001 0.553 0.459–0.666 Work 75.693 2 < 0.001 no problem vs. unable -0.899 0.114 61.785 1 < 0.001 0.407 0.325–0.509 short time vs. unable -0.138 0.108 1.633 1 0.201 0.871 0.705–1.076 Cognitive impairment 0.291 0.088 11.046 1 < 0.001 1.338 1.127–1.588 Deposit 0.504 0.116 18.742 1 < .0001 1.655 1.317–2.079 Table 3 Results of multivariate logistic regression analysis B S.E Wald df Sig. Exp(B) 95%C.I.for EXP(B) Health self-assessment 53.612 2 < 0.001 good vs. poor -0.985 0.171 33.025 1 < 0.001 0.373 0.267–0.522 common vs. poor -0.671 0.107 39.278 1 < 0.001 0.511 0.414–0.630 Pain 60.979 2 < 0.001 none vs. often -1.088 0.141 59.651 1 < 0.001 0.337 0.256–0.444 Occasionally vs. often -0.580 0.114 25.746 1 < 0.001 0.560 0.447-0.700 Sleep at night 36.563 2 < 0.001 8h 0.497 0.162 9.427 1 8h -0.110 0.161 0.469 1 0.494 0.895 0.653–1.228 Caregiver -0.597 0.104 32.638 1 < 0.001 0.551 0.449–0.676 Cognitive impairment 0.222 0.101 4.859 1 < 0.028 1.249 1.025–1.522 Predictive performance of disabled elderly Table 4 and Fig. 2 showed that except for decision trees, the use of LR and ML techniques for predicting depression among disabled elderly individuals is deemed acceptable. In the training set, XGBoost exhibited good predictive performance (AUC = 0.76). In the validation set, traditional LR showed better predictive performance (AUC = 0.73). Decision trees produced inconclusive results in both the training and validation sets (AUC < 0.70); however, the other models did not show significant differences in AUC. The overall predictive performance, as proved by the Brier score, demonstrated favorable outcomes. Figure 3 illustrated the calibration plot, showing good consistency between the predicted probabilities of LR and XGBoost and the actual observations. In Fig. 4 , within the threshold probability range of 0.15 to 0.89 for depression among disabled elderly individuals, implementing a selective intervention strategy yielded higher net gains than the default approach of intervention or non-intervention with all patients. Table 4 Comparison of Predictive Model Performance Model AUC Accuracy Sensitivity Specificity PPV NPV precision recall Brier-score Training set LR 0.73 0.67 0.79 0.53 0.69 0.66 0.74 0.59 0.20 RF 0.74 0.69 0.80 0.55 0.70 0.67 0.70 0.79 0.21 SVM 0.72 0.68 0.78 0.53 0.68 0.65 0.68 0.82 0.22 DT 0.67 0.67 0.79 0.52 0.68 0.65 0.68 0.82 0.22 XGB 0.76 0.69 0.78 0.58 0.71 0.67 0.71 0.79 0.20 Validation set LR 0.73 0.67 0.78 0.52 0.69 0.63 0.76 0.62 0.20 RF 0.72 0.66 0.76 0.51 0.68 0.61 0.73 0.79 0.21 SVM 0.72 0.67 0.78 0.53 0.69 0.63 0.69 0.83 0.22 DT 0.65 0.66 0.74 0.51 0.69 0.62 0.69 0.80 0.21 XGB 0.71 0.65 0.74 0.52 0.68 0.59 0.73 0.80 0.21 Visualization by SHAP We were using the SHAP algorithm to intuitively display the independent risk factors for predicting depression among functionally impaired older adults using the XGBOOST model. (Fig. 5 A, B) demonstrated the ranking of the importance of risk factors in descending order. Self-rated health has the most vital predictive ability, followed by pain and shorter sleep duration at night. (Fig. 5 C) SHAP provides feature interaction diagrams to identify features suitable for combination. We also provide two typical examples, one predicting no depression (Fig. 5 D) and the other predicting depression (Fig. 5 E), to demonstrate the interpretability of the model. Construction of nomogram Based on the logistic regression analysis results, a nomogram is constructed for the independent risk factors mentioned above. As shown in Fig. 6 , each axis represents a specific variable, and the corresponding values for each variable are found along the axis, typically marked with scales. Then, summing these values yields a total score corresponding to the predicted probability, indicating the higher the total score, the greater the likelihood of depression among disabled elderly individuals. Discussion We investigated the contributing factors between disability among elderly individuals and depression using cross-sectional data from a representative Chinese population. The predictive factors consisted of " health self-assessment", "pain", " sleep at night", "caregiver" and "cognitive impairment." This study was the first to predict the risk of depression in disabled individuals aged 60 years or older in China, and it also compared various machine learning techniques with traditional logistic regression. All models, except for decision trees, demonstrate acceptable discriminative capability; however, their performance falls short of the desired threshold (AUC ≥ 0.9) [ 33 ]. This may be because the concept of depressive tendencies involves multiple aspects and diagnostic approaches, and currently, there are no valuable biomarkers or biological screening tests in clinical practice [ 34 ]. Conversely, prediction models have demonstrated good performance in diseases characterized by well-defined conditions, such as stroke and diabetes [ 35 , 36 ]. It is worthwhile to consider some depression-related risk factors [ 34 ]. Therefore, it is of practical significance to predict depression in disabled elderly individuals by identifying crucial predictive factors [ 37 , 38 ]. Risk prediction models for depression, integrating variables such as cognitive assessments and self-rated health, achieve satisfactory performance when applied to cross-sectional data. The sensitivity values of all models (ranging from 0.74 to 0.80) are higher compared to the specificity values (ranging from 0.51 to 0.58). Greater sensitivity contributes to a lower false-negative rate, facilitating the identification of a larger proportion of elderly individuals exhibiting depressive tendencies. This enhances the awareness of primary healthcare workers toward disabled older adults with depressive tendencies and allows for early prevention strategies targeting this specific population group. Furthermore, the results of DCA indicate that all models can be used in clinical practice within a reasonable range of threshold probabilities[ 28 ]. Predictive models built using readily accessible variables will be employed in diverse scenarios for the identification of individuals at risk of depression and the implementation of preventive interventions[ 39 , 40 ]. "Poor self-rated health," "pain," " shorter sleep duration at night," " absence of caregivers," and "cognitive impairment" are high-risk factors for depression in disabled elderly individuals. The most prominent risk factor identified in this study is "poor self-rated health," which measures self-perceived health. Previous research has confirmed the correlation between self-rated health and depression.[ 41 , 42 ]. For individuals with impaired functioning, perceiving oneself as physically unhealthy is more likely to trigger severe depressive symptoms than perceiving oneself as physically healthy. This finding implies that the elderly population's perception of health may exert a more substantial influence on their depressive symptoms. Depression and pain share significant pathophysiological overlap, with a higher prevalence of depression observed in patients with chronic pain compared to those without pain[ 43 ]. Furthermore, the simultaneous presence of pain and depressive symptoms adversely affects individuals[ 44 ]. Over 60% of individuals with depression experience chronic pain symptoms[45]. The comorbidity between pain and depression holds particular significance in clinical settings, as the simultaneous presence of chronic pain and depression in older adults presents treatment challenges[ 46 ]. Furthermore, depression and pain have a bidirectional relationship where they can act as risk factors for each other[ 47 ]. Short sleep duration may increase daytime fatigue, which can lead to adverse events and emotions caused by fatigue, ultimately resulting in depression[ 48 ].In addition, research has shown that longer sleep duration is associated with lower levels of physical activity, which is beneficial for reducing the risk of depression[ 49 ]. Through elevation of neurotransmitter levels, specifically dopamine and serotonin, and enhancement of brain noradrenergic synaptic transmission[ 50 ], this mechanism promotes endorphin secretion[ 51 ], diminishes stress stimuli[ 52 ], and enhances self-efficacy and self-esteem. A longitudinal study revealed a robust correlation between depression scores and sleep duration, indicating a substantial impact of inadequate sleep on depression[ 53 ]. Furthermore, another prospective study showed that shorter sleep duration is associated with increased severity of depressive symptoms[ 54 ]. There is a strong bidirectional relationship between sleep and depression, where reduced sleep duration serves as the strongest predictor for increased acute depressive symptoms. Moreover, more than 80% of individuals with depression experience disturbances in their sleep patterns. Inadequate social support has a detrimental impact on mental health[ 55 ], whereas sufficient social support exerts a positive influence on the well-being of individuals experiencing depression. Previous studies have consistently identified spousal care for elderly individuals as a protective factor against depression. This is especially true for males, as poor marital relationships or the absence of a partner at home are related to depression in elderly individuals[ 56 ]. Caregivers can provide accommodation and meals, thereby reducing the need for hospitalization. They can also make more use of social networks and provide support by accompanying individuals during treatment. Insightful family members or close friends can serve as an "early warning system," enabling early intervention for individuals who may be at risk of depression[ 57 ]. Compared to elderly individuals with caregivers, empty nesters show a significantly higher tendency towards depression than non-empty nesters[ 58 ]. Studies have demonstrated a prevalence rate exceeding 70% for depression or depressive symptoms among empty nesters in China[ 59 , 60 ]. Depression may co-occur with or even precede dementia, which is characterized by diffuse cognitive impairment[ 61 ]. Studies have indicated an association between cognitive dysfunction and late-life depression, as well as adverse reactions to antidepressant medications[ 62 ]. Furthermore, cognitive impairment plays a significant role in the development of depression[ 63 ]. Cognitive impairments are observed in individuals at the onset of depression. Recurrent episodes of depression exhibit more significant cognitive impairments compared to single episodes, particularly in processing speed, executive function, language learning, and memory[ 64 ]. We have developed a risk prediction model for depression among disabled elderly individuals. By incorporating five high-risk factors: " health self-assessment", "pain", " sleep at night", "caregiver" and "cognitive impairment." We have developed a web-based clinical support system that is user-friendly and community-oriented. This system will facilitate the early identification of elderly individuals with depressive tendencies by themselves and caregivers, promoting proactive care and enhancing healthcare allocation through targeted interventions for disease prevention and management. Strengths and limitations To our knowledge, this is the first nomogram constructed based on the Chinese elderly population to predict depression among disabled older adults. This nomogram accurately identifies individuals with high-risks by incorporating selected independent risk factors. This tool will assist caregivers and healthcare practitioners in implementing timely intervention strategies to prevent depression among disabled older adults. Inevitably, this study has some limitations. Firstly, caution should be exercised when extrapolating the findings of this study to other countries as it is solely based on the Chinese population. Secondly, the presence of biases induced by missing data must be acknowledged. To ensure the robustness of the predictive model for depression among disabled elderly individuals, participants with missing data or abnormal values were excluded from the analysis. Thirdly, we could not obtain other important predictive factors, such as specific diets and physical activity, that were not collected in the CHARLS dataset[ 65 ]. Conclusion In the study, the logistic regression and XGBoost models demonstrated good discrimination, calibration, overall predictive performance, and clinical utility in predicting depression among disabled elderly individuals. A straightforward and efficient preliminary clinical support system was developed based on the logistic regression model, showing promise to significantly reduce the burden on users and help healthcare service providers manage depression. Abbreviations ADL activity of daily living IADL instrumental activity of daily living PSMS physical self-maintenance scale CESD-10 10-item Center for Epidemiologic Studies Depression Scale ROC receiver operating characteristic AUC area under the curve DCA decision curve analysis SHAP Shapley Additive exPlanations XGBoost eXtreme Gradient Boosting HDL-C high-density lipoprotein cholesterol LR logistic regression ML machine learning SVM support vector machine Declarations Acknowledgements We thank the Chinese Health and Retirement Longitudinal Study team for generously providing the required data. Additionally, our appreciation goes to the participating students, volunteers, and staff who contributed to this research endeavor. Authors’ contributions Shanshan Hong conceptualized and planned the overall study, conducted data analysis, and drafted the manuscript. Bingqian Lu,Yan Jiang and Shaobing Wang critically reviewed the manuscript. All authors read and approved the final manuscript. Fundings Faculty development grants from Hubei University of Medicine (2020QDJZR016) Wuhan Optoelectronics National Research Center 2021 Open Fund Project (2021WNLOKF020). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Data Availability The database was used from the China Health and Retirement Longitudinal Study. (http://charls.pku.edu.cn/). Ethics approval and consent to participate This study was conducted following the principles of the Helsinki Declaration and was approved by the Biomedical Ethics Committee of Peking University. All participants signed informed consent forms before their participation, which were approved by the Ethics Review Committee of Peking University (IRB00001052-11015). Consent for publication Not applicable. Competing interests The authors declare no competing interests. References Zheng X, Pang L, Chen G, Huang C, Liu L, Zhang L. Challenge of Population Aging on Health. Public Health Challenges in Contemporary Ch-ina: An Interdisciplinary Perspective.2016:43–53. https://doi.org/10.1007/978-3-662-47753-3_4 . Stordal E, Mykletun A, Dahl AA. The association between age and depression in the general population: a multivariate examination. Acta psychiatrica Scandinavica. 2003;107:132–41. Whiteford HA, Ferrari AJ, Degenhardt L, Feigin V, Vos T. The Global Burden of Mental, Neurological and Substance Use Disorders: An Analysis from the Global Burden of Disease Study 2010. PLoS ONE. 2015;10:14. Mueller TI, Kohn R, Leventhal N, Leon AC, Solomon D, Coryell W, Endicott J, Alexopoulos GS, Keller MB. The course of depression in elderly patients. Am J Geriatr Psychiatry. 2004;12:22–9. Mitchell AJ, Subramaniam H. Prognosis of depression in old age compared to middle age: a systematic review of comparative studies. Am J Psychiatry. 2005;162:1588–601. Verhaak PF, Dekker JH, de Waal MW, van Marwijk HW. Comijs HC. Depression, disability and somatic diseases among elderly. J Affect Disord. 2014;167:187–91. Robinson-Whelen S, Taylor HB, Hughes RB, Wenzel L, Nosek MA. Depression and depression treatment in women with spinal cord injury. Top Spinal Cord Inj Rehabil. 2014;20:23–31. Wanjau MN, Möller H, Haigh F, Milat A, Hayek R, Lucas P, Veerman JL. Physical Activity and Depression and Anxiety Disorders in Australia: A Lifetable Analysis. AJPM Focus. 2023;2:100030. Noh JW, Kwon YD, Park J, Oh IH, Kim J. Relationship between Physical Disability and Depression by Gender: A Panel Regression Model. PLoS ONE. 2016;11:e0166238. Djernes JK. Prevalence and predictors of depression in populations of elderly: a review. Acta Psychiatr Scand. 2006;113:372–87. Hassler AP, Menasalvas E, García-García FJ, Rodríguez-Mañas L, Holzinger A. Importance of medical data preprocessing in predictive modeling and risk factor discovery for the frailty syndrome. BMC Med Inform Decis Mak. 2019;19:33. Akbari G, Nikkhoo M, Wang L, Chen CPC, Han DS, Lin YH, Chen HB, Cheng CH. Frailty Level Classification of the Community Elderly Using Microsoft Kinect-Based Skeleton Pose: A Machine Learning Approach. Sens (Basel). 2021;21. 10.3390/s21124017 . Bakhtiari A, Hashemi M, Hosseini SR, Omidvar S, Bijani A, Khairkhah F. The Relationship between Depression and Metabolic Syndrome in the Elderly Population: The Cohort Aging Study. Iran J Psychiatry. 2018;13:230–8. Tyler Williamson P, Aponte-Hao S, Mele B, Lethebe BC, Leduc C, Thandi M, Katz A, Wong ST. Developing and Validating a Primary Care EMR-based Frailty Definition using Machine Learning. Int J Popul Data Sci. 2020;5:1344. Park C, Mishra R, Sharafkhaneh A, Bryant MS, Nguyen C, Torres I, Naik AD, Najafi B. Digital Biomarker Representing Frailty Phenotypes: The Use of Machine Learning and Sensor-Based Sit-to-Stand Test. Sens (Basel). 2021;21. 10.3390/s21093258 . Shin C, Park MH, Lee SH, Ko YH, Kim YK, Han KM, Jeong HG, Han C. Usefulness of the 15-item geriatric depression scale (GDS-15) for classifying minor and major depressive disorders among community-dwelling elders. J Affect Disord. 2019;259:370–5. Zhao Y, Hu Y, Smith JP, Strauss J, Yang G. Cohort profile: the China Health and Retirement Longitudinal Study (CHARLS). Int J Epidemiol. 2014;43:61–8. Katz S, Ford AB, Moskowitz RW, Jackson BA, Jaffe MW, STUDIES OF ILLNESS IN THE AGED. THE INDEX OF ADL: A STANDARDIZED MEASURE OF BIOLOGICAL AND PSYCHOSOCIAL FUNCTION. JAMA. 1963;185:914–9. Covinsky KE, Hilton J, Lindquist K, Dudley RA. Development and validation of an index to predict activity of daily living dependence in community-dwelling elders. Med Care. 2006;44:149–57. Kingston A, Comas-Herrera A, Jagger C. Forecasting the care needs of the older population in England over the next 20 years: estimates from the Population Ageing and Care Simulation (PACSim) modelling study. Lancet Public Health. 2018;3:e447–55. Zhou L, Ma X, Wang W. Relationship between Cognitive Performance and Depressive Symptoms in Chinese Older Adults: The China Health and Retirement Longitudinal Study (CHARLS). J Affect Disord. 2021;281:454–8. Dong L, Freedman VA, Mendes de Leon CF. The Association of Comorbid Depression and Anxiety Symptoms With Disability Onset in Older Adults. Psychosom Med. 2020;82:158–64. Chen CM, Liu LF. The effect of disability and depression on cognitive function and screening factors. Arch Gerontol Geriatr. 2017;73:154–9. Gortmaker SL, Hosmer DW, Lemeshow SJCS. Applied Logistic Regression. 3rd ed. Hoboken: John Wiley & Sons; 2013. Zhang H, Chen D, Shao J, Zou P, Cui N, Tang L, Wang X, Wang D, Ye Z. External Validation of the Prognostic Prediction Model for 4-Year Risk of Metabolic Syndrome in Adults: A Retrospective Cohort Study. Diabetes Metab Syndr Obes. 2021;14:3027–34. Huang YQ, Liang CH, He L, Tian J, Liang CS, Chen X, Ma ZL, Liu ZY. Development and Validation of a Radiomics Nomogram for Preoperative Prediction of Lymph Node Metastasis in Colorectal Cancer. J Clin Oncol. 2016;34:2157–64. Steyerberg EW, Vickers AJ, Cook NR, Gerds T, Gonen M, Obuchowski N, Pencina MJ, Kattan MW. Assessing the performance of prediction models: a framework for traditional and novel measures. Epidemiology.2010;21:128 – 38. Kerr KF, Brown MD, Zhu K, Janes H. Assessing the Clinical Impact of Risk Prediction Models With Decision Curves: Guidance for Correct Interpretation and Appropriate Use. J Clin Oncol. 2016;34:2534–40. Lundberg S, Lee S-I. A unifed approach to interpreting model predictions.ArXiv170507874 Cs Stat. 2017 [cited 1 Sep 2020]. Available: http://arxiv.org/abs/1705.07874 . 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. 2020 IEEE 20th International Conference on Bioinformatics and Bioengineering (BIBE).2020;859–864. Joseph AJapa. Parametric inference with universal function approximators. arXiv preprint arXiv: 1903.04209 ML Stat. 2019 [cited 4 Oct 2020]. Available: https://doi.org/10.48550/arXiv.1903.04209 . Lundberg SM, Nair B, Vavilala MS, Horibe M, Eisses MJ, Adams T, Liston DE, Low DK-W. Newman S-F and Kim JJNbe. Explainable machine-learning predictions for the prevention of hypoxaemia during surgery. 2018;2:749–60. Wang QQ, Yu SC, Qi X, Hu YH, Zheng WJ, Shi JX, Yao HY. [Overview of logistic regression model analysis and application]. Zhonghua Yu Fang Yi Xue Za Zhi. 2019;53:955–60. McCarron RM, Shapiro B, Rawles J, Luo. J Depress Ann Intern Med. 2021;174:Itc65–itc80. Park E, Chang HJ, Nam HS. A Bayesian Network Model for Predicting Post-stroke Outcomes With Available Risk Factors. Front Neurol. 2018;9:699. Ting DSW, Cheung CY, Lim G, Tan GSW, Quang ND, Gan A, Hamzah H, Garcia-Franco R, San Yeo IY, Lee SY, Wong EYM, Sabanayagam C, Baskaran M, Ibrahim F, Tan NC, Finkelstein EA, Lamoureux EL, Wong IY, Bressler NM, Sivaprasad S, Varma R, Jonas JB, He MG, Cheng CY, Cheung GCM, Aung T, Hsu W, Lee ML, Wong TY. Development and Validation of a Deep Learning System for Diabetic Retinopathy and Related Eye Diseases Using Retinal Images From Multiethnic Populations With Diabetes. JAMA. 2017;318:2211–23. Brown CS. Depression and anxiety disorders. Obstet Gynecol Clin North Am. 2001;28:241–68. Fried EI, Epskamp S, Nesse RM, Tuerlinckx F, Borsboom D. What are 'good' depression symptoms? Comparing the centrality of DSM and non-DSM symptoms of depression in a network analysis. J Affect Disord. 2016;189:314–20. Tarekegn A, Ricceri F, Costa G, Ferracin E, Giacobini M. Predictive Modeling for Frailty Conditions in Elderly People: Machine Learning Approaches. JMIR Med Inform. 2020;8:e16678. Dong BR, Gu XQ, Chen HY, Gu J, Pan ZG. Development and Validation of a Nomogram to Predict Frailty Progression in Nonfrail Chinese Community-Living Older Adults. J Am Med Dir Assoc. 2021;22:2571–2578e4. Jahn DR, Cukrowicz KC. Self-rated health as a moderator of the relation between functional impairment and depressive symptoms in older adults. Aging Ment Health. 2012;16:281–7. Lyness JM, King DA, Conwell Y, Duberstein PR, Eberly S, Sörensen SM, Caine ED. Self-rated health, depression, and one-year health outcomes in older primary care patients. Am J Geriatr Psychiatry. 2004;12:110–3. Depression and pain. J Clin Psychiatry. 2008;69:1970–8. Robinson MJ, Edwards SE, Iyengar S, Bymaster F, Clark M, Katon W. Depression and pain. Front Biosci (Landmark Ed).2009;14:5031-51.Chronic pain and major depressive disorder in the general population. J Psychiatr Res. 2010;44:454–61. Fishbain DA, Cutler R, Rosomoff HL, Rosomoff RS. Chronic pain-associated depression: antecedent or consequence of chronic pain? A review. Clin J Pain. 1997;13:116–37. Zis P, Daskalaki A, Bountouni I, Sykioti P, Varrassi G, Paladini A. Depression and chronic pain in the elderly: links and management challenges. Clin Interv Aging. 2017;12:709–20. Zhai L, Zhang H, Zhang D, SLEEP DURATION AND, DEPRESSION. AMONG ADULTS: A META-ANALYSIS OF PROSPECTIVE STUDIES. Depress Anxiety. 2015;32:664–70. Stranges S, Dorn JM, Shipley MJ, Kandala NB, Trevisan M, Miller MA, Donahue RP, Hovey KM, Ferrie JE, Marmot MG, Cappuccio FP. Correlates of short and long sleep duration: a cross-cultural comparison between the United Kingdom and the United States: the Whitehall II Study and the Western New York Health Study. Am J Epidemiol. 2008;168:1353–64. Weicker H, Strüder HK. Influence of exercise on serotonergic neuromodulation in the brain. Amino Acids. 2001;20:35–47. Janal MN, Colt EWD, Clark CW, Glusman M. Pain sensitivity, mood and plasma endocrine levels in man following long-distance running: effects of naloxone. Pain. 1984;19:13–25. Salmon P. Effects of physical exercise on anxiety, depression, and sensitivity to stress: a unifying theory. Clin Psychol Rev. 2001;21:33–61. Firth-Cozens J. Individual and organizational predictors of depression in general practitioners. Br J Gen Pract. 1998;48:1647–51. Buckley T, Bartrop R, McKinley S, Ward C, Bramwell M, Roche D, Mihailidou AS, Morel-Kopp MC, Spinaze M, Hocking B, Goldston K, Tennant C, Tofler G. Prospective study of early bereavement on psychological and behavioural cardiac risk factors. Intern Med J. 2009;39:370–8. Brown GW, Andrews B. Social support and depression. Dynamics of stress: Physiological, psychological and social perspectives.1986;257–282. Sonnenberg CM, Deeg DJ, van Tilburg TG, Vink D, Stek ML, Beekman AT. Gender differences in the relation between depression and social support in later life. Int Psychogeriatr. 2013;25:61–70. Smith L, Hill N, Kokanovic R. Experiences of depression, the role of social support and its impact on health outcomes. J Ment Health. 2015;24:342–6. Zhai Y, Yi H, Shen W, Xiao Y, Fan H, He F, Li F, Wang X, Shang X, Lin J. Association of empty nest with depressive symptom in a Chinese elderly population: A cross-sectional study. J Affect Disord. 2015;187:218–23. Xie LQ, Zhang JP, Peng F, Jiao NN. Prevalence and related influencing factors of depressive symptoms for empty-nest elderly living in the rural area of YongZhou, China. Arch Gerontol Geriatr. 2010;50:24–9. Su D, Wu XN, Zhang YX, Li HP, Wang WL, Zhang JP, Zhou LS. Depression and social support between China' rural and urban empty-nest elderly. Arch Gerontol Geriatr. 2012;55:564–9. Kiosses DN, Alexopoulos GS. IADL functions, cognitive deficits, and severity of depression: a preliminary study. Am J Geriatr Psychiatry. 2005;13:244–9. Morimoto SS, Kanellopoulos D, Manning KJ, Alexopoulos GS. Diagnosis and treatment of depression and cognitive impairment in late life. Ann N Y Acad Sci. 2015;1345:36–46. Grahek I, Shenhav A, Musslick S, Krebs RM, Koster EHW. Motivation and cognitive control in depression. Neurosci Biobehav Rev. 2019;102:371–81. Varghese S, Frey BN, Schneider MA, Kapczinski F, de Azevedo Cardoso T. Functional and cognitive impairment in the first episode of depression: A systematic review. Acta Psychiatr Scand. 2022;145:156–85. Peng LN, Hsiao FY, Lee WJ, Huang ST, Chen LK. Comparisons Between Hypothesis- and Data-Driven Approaches for Multimorbidity Frailty Index: A Machine Learning Approach. J Med Internet Res. 2020;22:e16213. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 14 Feb, 2025 Read the published version in BMC Psychiatry → Version 1 posted Editorial decision: Revision requested 20 Jul, 2024 Reviews received at journal 06 Jul, 2024 Reviews received at journal 30 Jun, 2024 Reviewers agreed at journal 30 Jun, 2024 Reviewers agreed at journal 26 Jun, 2024 Reviewers agreed at journal 26 Jun, 2024 Reviewers agreed at journal 25 May, 2024 Reviewers invited by journal 02 May, 2024 Editor assigned by journal 30 Apr, 2024 Editor invited by journal 22 Jan, 2024 Submission checks completed at journal 22 Jan, 2024 First submitted to journal 18 Jan, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3874875","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":268489695,"identity":"1e88e8e0-71ab-45d5-b237-2e7a2d182ba5","order_by":0,"name":"Shanshan Hong","email":"","orcid":"","institution":"Hubei University of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Shanshan","middleName":"","lastName":"Hong","suffix":""},{"id":268489696,"identity":"5a9e1ee8-06c1-41a6-8757-04c334bb4a19","order_by":1,"name":"Bingqian Lu","email":"","orcid":"","institution":"Hubei University of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Bingqian","middleName":"","lastName":"Lu","suffix":""},{"id":268489697,"identity":"53b3d8f6-6cd1-4faa-8aef-3f934c89ba6c","order_by":2,"name":"Shaobing Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAu0lEQVRIiWNgGAWjYHACZoYPBjZybOztB4jXwjijIM2Yj+dMAvFamHk+HE6cJ+FgQJx6+f6zh415DJjT2yQYEhh+VGwjrMXgRl5y4hwDttw26cYDjD1nbhOhRYLH+MAbA57cNpkDCcyMbURoke8/Y3yAx0AinU0iwYA4LQwHcowTeQwMEojXAvKL4QyDBMM2YCAfJMovoBCT+PDnv7x8e/vBBz8qiHEYAw+SI4lRj6plFIyCUTAKRgFWAAAthzoWoSp0MAAAAABJRU5ErkJggg==","orcid":"","institution":"Hubei University of Medicine","correspondingAuthor":true,"prefix":"","firstName":"Shaobing","middleName":"","lastName":"Wang","suffix":""},{"id":268489698,"identity":"675b3bd6-75c3-4f7b-8c89-a0aadcf30217","order_by":3,"name":"Yan Jiang","email":"","orcid":"","institution":"Hubei University of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Yan","middleName":"","lastName":"Jiang","suffix":""}],"badges":[],"createdAt":"2024-01-18 06:30:07","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3874875/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3874875/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12888-025-06577-x","type":"published","date":"2025-02-14T15:57:53+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":50117431,"identity":"3b9c58d5-d356-47fc-ab06-38b12e39c778","added_by":"auto","created_at":"2024-01-24 18:57:23","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":300947,"visible":true,"origin":"","legend":"\u003cp\u003eParticipant selection flowchart in this study\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-3874875/v1/60489aad46874db5d562105b.jpeg"},{"id":50117434,"identity":"85da0f68-52f9-47fe-86bf-524bad5df10a","added_by":"auto","created_at":"2024-01-24 18:57:23","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":414248,"visible":true,"origin":"","legend":"\u003cp\u003eIn the context of training set (A) and validation set (B), ROC curves were constructed to illustrate the predictive performance of the top five variables for depression among disabled elderly individuals using five distinct models. The x-axis denotes specificity, defined as the probability of a negative detection when depression is absent in disabled elderly individuals, while the y-axis represents sensitivity, indicating the probability of a positive detection when depression is present in disabled elderly individuals.\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-3874875/v1/710a02663198864a1fd8ae85.jpeg"},{"id":50117435,"identity":"ffb86b19-ffaf-4afd-9f26-12752bb293e4","added_by":"auto","created_at":"2024-01-24 18:57:23","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":576704,"visible":true,"origin":"","legend":"\u003cp\u003eThe figure below depicts the calibration curves for the training set \u003cstrong\u003e(A)\u003c/strong\u003e and validation set (B).\u003c/p\u003e","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-3874875/v1/e431e4326361a4d21dd1d3d7.jpeg"},{"id":50117432,"identity":"4e6ecc22-164a-425a-a757-592d2453a022","added_by":"auto","created_at":"2024-01-24 18:57:23","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":503438,"visible":true,"origin":"","legend":"\u003cp\u003eThe following figure illustrates the decision curve analysis of five predictive models for depression in disabled elderly individuals within the training set\u003cstrong\u003e(A)\u003c/strong\u003e and validation set \u003cstrong\u003e(B)\u003c/strong\u003e. The x-axis represents the threshold probability for depression, while the y-axis denotes the net benefit.\u003c/p\u003e","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-3874875/v1/8b8604f380ad991a53cdedde.jpeg"},{"id":50118008,"identity":"af7ed33e-c820-4826-a3d4-792392e8c11d","added_by":"auto","created_at":"2024-01-24 19:05:23","extension":"jpeg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":585003,"visible":true,"origin":"","legend":"\u003cp\u003eSHAP summary plot for the identified independent risk factors contributing to the XGBoost model. (\u003cstrong\u003eA)\u003c/strong\u003e In the figure, different colors represent levels of variable values—yellow indicates higher levels, while purple indicates lower levels. The thickness of the line, which is composed of individual dots, corresponds to the number of samples at a given value. The x-axis represents the influence of the variable on the outcome, where a positive SHAP value indicates an increase in risk, and a negative SHAP value indicates a decrease in risk. (\u003cstrong\u003eB) \u003c/strong\u003eSHAP feature importance is measured as the mean absolute Shapley values. This matrix chart illustrates the importance of independent risk factors in the development of the XGBoost model. (\u003cstrong\u003eC)\u003c/strong\u003eSHAP provides feature interaction plots to identify features suitable for combinations. Features highlighted in yellow and purple in the plot indicate that constructing cross-features can effectively enhance the model's performance. \u003cstrong\u003e(D) \u003c/strong\u003eSHAP prediction without depression. \u003cstrong\u003e(E)\u003c/strong\u003e SHAP prediction with depression. Yellow arrows indicate a higher risk of depression, while purple arrows indicate a lower risk of depression. The length of the arrows helps visualize the degree of influence of the features, so the longer the arrow, the more important the feature is to the outcome.\u003c/p\u003e","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-3874875/v1/79c2ff3c1fd154be64565464.jpeg"},{"id":50117430,"identity":"4e49f52d-8cfb-400a-a409-8107f4d0ca7d","added_by":"auto","created_at":"2024-01-24 18:57:23","extension":"jpeg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":73529,"visible":true,"origin":"","legend":"\u003cp\u003eA nomogram was developed within the training dataset, incorporating health self-assessment, pain, caregiver, cognitive impairment, and sleep at night.\u003c/p\u003e","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-3874875/v1/b56d37b2323996c2a819644b.jpeg"},{"id":76487565,"identity":"2013fa36-158b-4979-bcf4-ad5e3ef56c89","added_by":"auto","created_at":"2025-02-17 16:09:13","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3610302,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3874875/v1/7f081fdb-853c-44ac-a5b3-115527a78039.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Comparison of logistic regression and machine learning methods for predicting depression risks among disabled elderly individuals: results from the China Health and Retirement Longitudinal Study","fulltext":[{"header":"Background","content":"\u003cp\u003eResearch indicates that the population of disabled elderly individuals in China has exceeded 40\u0026nbsp;million, making up over 16% of the elderly population. Projections anticipate that the number will reach 65\u0026nbsp;million by 2030 [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Depression is a prevalent mental health disorder among the elderly population, impacting around 7% of them worldwide [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Both disability and depression have emerged as notable social concerns. Disability encompasses physical, psychological, and social aspects and pertains to the impaired capacity of older individuals to engage in fundamental activities of daily living autonomously. This decline in functional abilities hampers their adaptation to environmental changes. Depression is a common psychological issue characterized by a chronic trajectory and a high likelihood of recurrence [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. It is associated with various adverse outcomes, such as reduced quality of life, amplified healthcare burden, and heightened incidence and mortality rates [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Both disability and depression impose substantial burdens on individuals as well as society.\u003c/p\u003e \u003cp\u003eLate-life depression is closely associated with disability, particularly when it hinders self-care and social participation [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Research indicates that the likelihood of depression in the population with physical disabilities is at least three times higher compared to normal individuals [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Moreover, there is a potential direct causal link between disability among older adults and the occurrence of depression. There could exist a direct causal relationship between disability and depression in the older adult population [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e].\u003c/p\u003e \u003cp\u003ePrevious research has indicated that several significant factors are associated with the onset and persistence of depression in older adults, including being female, having a low educational level, experiencing spousal loss, cognitive decline, physical illness, and functional impairment [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Currently, the relationship between disability in elderly individuals and depression is still the subject of ongoing investigations, and interventions for depression among disabled older adults are considered essential for healthcare services.\u003c/p\u003e \u003cp\u003eClinical risk prediction models have become widely prevalent in various medical fields in recent years. These models aim to utilize individual-level information to predict clinically relevant outcomes [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. By employing clinical prediction models, the efficiency of healthcare processes and self-management can be enhanced while reducing the workforce costs associated with healthcare [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn certain studies, relevant biochemical indicators such as HDL-C, fasting blood glucose, triglycerides, and other metabolic markers are integrated into prediction models [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. However, the process of collecting data for these indicators is invasive or costly, necessitating well-trained experts and controlled experimental environments, often rendering them impractical in community settings. While the inclusion of biochemical indicators may enhance the predictive performance of risk models, it may also diminish their practicality [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Primary healthcare demands more straightforward tools [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. In primary healthcare settings, risk prediction models based on readily accessible data concerning risk factors associated with depression in disabled elderly individuals prove to be more suitable and feasible [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWe constructed a clinical prediction model for depression among disabled elderly individuals by analyzing data from the 2018 China Health and Retirement Longitudinal Study (CHARLS) to identify the relevant factors associated with depression in this population. By utilizing commonly available predictive factors in the community environment, we developed a practical prediction system for assessing the risks of depression among disabled elderly individuals. This system aims to assist healthcare professionals in measuring the probability of depression in this population and facilitate to identify the risk of depression at an early stage among these individuals.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy design and participants\u003c/h2\u003e \u003cp\u003eThe data was derived from the China Health and Retirement Longitudinal Study (CHARLS). CHARLS is conducted every two years (2011, 2013, 2015, 2018), with each wave adding new participants. The CHARLS participants were sampled using a multistage probability sampling strategy and probability proportionate to the size sampling method. It covered 150 counties of 28 provinces, municipal cities, and autonomous regions of China [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. The CHARLS was approved by the Ethical Review Committee at Peking University, and all participants signed informed consent before participation.\u003c/p\u003e \u003cp\u003eWe employed the baseline data from 2018 to develop a practical risk prediction model for depressive symptoms among disabled older adults. In the longitudinal CHARLS cohort, disability is classified into two dimensions: Activities of Daily Living (ADL) and Instrumental Activities of Daily Living (IADL). To assess disability, this study utilized the PSMS scale and Lawton scale, each comprising six sections[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. In the present study, participants who acknowledge difficulties by responding with \u0026ldquo;yes, I have difficulty and need assistance\u0026rdquo;, or express an inability with \u0026ldquo;I cannot do it\u0026rdquo; in the context of any given project, are regarded as functionally impaired[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. Participants were considered disabled if any section was classified as such. The 2018 baseline survey included a total of 19,817 participants. After excluding individuals with missing or abnormal data, the final analytical sample consisted of 3,107 disabled older adults aged 60 years and above. Figure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e illustrates the flowchart for selecting and following up on all eligible study subjects.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eMeasures\u003c/h2\u003e \u003cdiv id=\"Sec5\" class=\"Section3\"\u003e \u003ch2\u003eDepressive symptoms\u003c/h2\u003e \u003cp\u003eIn the CHARLS cohort, depression is measured by using the 10-item Center for Epidemiologic Studies Depression Scale (CESD-10) administered through the survey [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. This scale consists of 10 components: \"feeling down,\" \"feeling afraid,\" \"feeling lonely,\" \"having poor sleep,\" \"having difficulty in completing tasks,\" \"being irritable over minor matters,\" \"having trouble concentrating\" and \"feeling unable to continue with life.\" The items \"I have hope for the future\" and \"I feel happy\" in this scale are scored inversely. Each question offers four response options: Rarely or none of the time (\u0026lt;\u0026thinsp;1 day), Some or a little of the time (1\u0026ndash;2 days), Occasionally or a moderate amount of the time (3\u0026ndash;4 days), and most or all of the time (5\u0026ndash;7 days).The overall scale score ranges from 0 to 30, with a threshold of 10 points indicating a tendency towards depression.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section3\"\u003e \u003ch2\u003eVariables\u003c/h2\u003e \u003cp\u003eConsidering the accuracy and practicality of the risk prediction model, we have incorporated variables related to depression among disabled older adults based on a review of existing evidence [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. Relevant information was collected through structured questionnaires, including sociodemographic factors (age, gender, place of residence, educational level, marital status, etc), health behaviors (smoking, drinking history, sleep, etc.), cognitive status and social support.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eData cleaning, preprocessing and merging were conducted using the STATA 17.0 software. The analysis of influencing factors and determining important predictive factors were performed on the sample using SPSS 27.0. The chi-square test for categorical variables was employed to compare the influencing factors of depression among disabled older adults and select variables with statistical significance. Important predictive factors related to depression among disabled older adults were selected based on expert opinions and through univariate and multivariate logistic regression analysis using the forward stepwise selection method. The statistical significance level was set at 0.05 for two-tailed tests.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eModel construction and evaluation\u003c/h2\u003e \u003cp\u003eA depression risk prediction model was constructed for elderly individuals with disabilities through the utilization of logistic regression and four machine learning algorithms: Support Vector Machine (SVM), Random Forest (RF), XGBoost, and Decision Tree. During the modeling process, conducting 10-fold cross-validation ensures the performance and stability of the model. The model's development was facilitated using R version 4.3.0.\u003c/p\u003e \u003cp\u003eCreate the receiver operating characteristic (ROC) curve to evaluate the predictive capability of the prediction model, where an area under the curve (AUC) value of \u0026ge;\u0026thinsp;0.70 is considered appropriate [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. The calibration curve visualizes the consistency assessment between the average predicted risk occurrence rate and the actual observed event occurrence rate. If the solid line (representing the performance of the predictive model) is closer to the diagonal dashed line (representing the perfect prediction of an ideal model), it indicates better consistency [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. A Brier score of less than 0.25 indicates that the overall performance evaluation of discrimination and calibration is acceptable [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. The decision curve evaluates the clinical utility by assessing the net benefit at different threshold probabilities [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e].\u003c/p\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003eModel interpretation\u003c/h2\u003e \u003cp\u003eSHAP, an acronym for Shapley Additive exPlanations, was introduced by Lundberg and Lee in 2017[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. It constitutes a framework employed for explicating the predictions made by machine learning models to elucidate each feature's contribution[\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. Shapley values, originating from cooperative game theory, serve as an equitable approach for allocating gains in cooperative games. About participants in cooperative games, Shapley values contemplate the marginal contributions of each participant, thereby ensuring a judicious distribution of benefits[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. In machine learning, features can be construed as participants in a cooperative game, with the model's output representing the dividends of the game[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eParticipant characteristics\u003c/h2\u003e \u003cp\u003eA total of 3,107 disabled elderly individuals were included in this study. Among them, 1,774 (57.1%) had depression, making up 57.1% of the total sample of disabled elderly individuals. Within the training set, 56.66% (1233/2176) of individuals exhibited depression, whereas in the validation set, the proportion was 58.11% (541/931). Characteristics of the study population are given in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCharacteristics of the participants at baseline\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTotal\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNon-depressive\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003edepression\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ex\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003ep\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003en\u0026thinsp;=\u0026thinsp;3107\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003en\u0026thinsp;=\u0026thinsp;1333\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003en\u0026thinsp;=\u0026thinsp;1774\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGender (%)\u003c/p\u003e \u003cp\u003emale\u003c/p\u003e \u003cp\u003efemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1278(41.13)\u003c/p\u003e \u003cp\u003e1829(58.87)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e623(46.74)\u003c/p\u003e \u003cp\u003e710(53.26)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e655(36.9)\u003c/p\u003e \u003cp\u003e1119(63.08)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e30.28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge, years\u003c/p\u003e \u003cp\u003e60\u0026ndash;69\u003c/p\u003e \u003cp\u003e70\u0026ndash;79\u003c/p\u003e \u003cp\u003e\u0026ge;\u0026thinsp;80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1726(55.55)\u003c/p\u003e \u003cp\u003e1101(35.44)\u003c/p\u003e \u003cp\u003e280(9.01)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e711(53.34)\u003c/p\u003e \u003cp\u003e476(35.71)\u003c/p\u003e \u003cp\u003e146(10.95)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1015(57.22)\u003c/p\u003e \u003cp\u003e625(35.23)\u003c/p\u003e \u003cp\u003e134(7.55)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e11.867\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eResidence\u003c/p\u003e \u003cp\u003evillage\u003c/p\u003e \u003cp\u003enon-rural\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2460(79.18)\u003c/p\u003e \u003cp\u003e647(20.82)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1013(75.99)\u003c/p\u003e \u003cp\u003e320(24.01)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1447(81.57)\u003c/p\u003e \u003cp\u003e327(18.43)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e14.34\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducation level\u003c/p\u003e \u003cp\u003eilliteracy\u003c/p\u003e \u003cp\u003eelementary school level\u003c/p\u003e \u003cp\u003ejunior high school level\u003c/p\u003e \u003cp\u003ehigh school and above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1143(36.79)\u003c/p\u003e \u003cp\u003e1429(46.00)\u003c/p\u003e \u003cp\u003e405(13.04)\u003c/p\u003e \u003cp\u003e130(4.18)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e462(34.66)\u003c/p\u003e \u003cp\u003e597(44.79)\u003c/p\u003e \u003cp\u003e198(14.85)\u003c/p\u003e \u003cp\u003e76(5.70)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e681(38.39)\u003c/p\u003e \u003cp\u003e832(46.90)\u003c/p\u003e \u003cp\u003e207(11.67)\u003c/p\u003e \u003cp\u003e54(3.04)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e22.386\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHealth self-assessment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003egood\u003c/p\u003e \u003cp\u003ecommon\u003c/p\u003e \u003cp\u003epoor\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e340(10.94)\u003c/p\u003e \u003cp\u003e1238(39.85)\u003c/p\u003e \u003cp\u003e1529(49.21)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e234(17.55)\u003c/p\u003e \u003cp\u003e638(47.86)\u003c/p\u003e \u003cp\u003e461(34.58)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e106(5.98)\u003c/p\u003e \u003cp\u003e600(33.82)\u003c/p\u003e \u003cp\u003e1068(60.20)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e232.416\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003epains\u003c/p\u003e \u003cp\u003enone\u003c/p\u003e \u003cp\u003eoccasionally\u003c/p\u003e \u003cp\u003eoften\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e668(21.50)\u003c/p\u003e \u003cp\u003e1339(43.10)\u003c/p\u003e \u003cp\u003e1100(35.40)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e424(31.81)\u003c/p\u003e \u003cp\u003e619(46.44)\u003c/p\u003e \u003cp\u003e290(21.76)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e244(13.75)\u003c/p\u003e \u003cp\u003e720(40.59)\u003c/p\u003e \u003cp\u003e810(45.66)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e244.266\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSleep at night\u003c/p\u003e \u003cp\u003e\u0026lt;\u0026thinsp;6h\u003c/p\u003e \u003cp\u003e6-8h\u003c/p\u003e \u003cp\u003e\u0026gt;\u0026thinsp;8h\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1454(46.80)\u003c/p\u003e \u003cp\u003e1328(42.74)\u003c/p\u003e \u003cp\u003e325(10.46)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e462(34.66)\u003c/p\u003e \u003cp\u003e707(53.04)\u003c/p\u003e \u003cp\u003e164(12.30)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e992(55.92)\u003c/p\u003e \u003cp\u003e621(35.01)\u003c/p\u003e \u003cp\u003e161(9.08)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e138.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSocial activities\u003c/p\u003e \u003cp\u003eno\u003c/p\u003e \u003cp\u003eyes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1689(54.36)\u003c/p\u003e \u003cp\u003e1418(45.64)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e680(51.01)\u003c/p\u003e \u003cp\u003e653(48.99)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1009(56.88)\u003c/p\u003e \u003cp\u003e765(43.12)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e10.55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCaregiver\u003c/p\u003e \u003cp\u003eyes\u003c/p\u003e \u003cp\u003eno\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2094(67.40)\u003c/p\u003e \u003cp\u003e1013(32.60)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1004(75.32)\u003c/p\u003e \u003cp\u003e329(24.68)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1090(61.44)\u003c/p\u003e \u003cp\u003e684(38.56)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e66.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWork\u003c/p\u003e \u003cp\u003eno problem\u003c/p\u003e \u003cp\u003eshort time\u003c/p\u003e \u003cp\u003eunable\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e965(31.06)\u003c/p\u003e \u003cp\u003e1250(40.23)\u003c/p\u003e \u003cp\u003e892(28.71)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e541(40.59)\u003c/p\u003e \u003cp\u003e477(35.78)\u003c/p\u003e \u003cp\u003e315(23.63)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e424(23.90)\u003c/p\u003e \u003cp\u003e773(43.57)\u003c/p\u003e \u003cp\u003e577(32.53)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e100.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCognitive impairment\u003c/p\u003e \u003cp\u003eno\u003c/p\u003e \u003cp\u003eyes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1357(43.68)\u003c/p\u003e \u003cp\u003e1750(56.32)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e644(48.31)\u003c/p\u003e \u003cp\u003e689(51.69)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e713(40.19)\u003c/p\u003e \u003cp\u003e1061(59.81)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e20.402\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDeposit\u003c/p\u003e \u003cp\u003e\u0026le;\u0026thinsp;5000\u003c/p\u003e \u003cp\u003e\u0026gt;\u0026thinsp;5000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2615(84.16)\u003c/p\u003e \u003cp\u003e492(15.84)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1068(80.12)\u003c/p\u003e \u003cp\u003e265(19.88)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1547(87.20)\u003c/p\u003e \u003cp\u003e227(12.80)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e28.658\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eVariable selection results\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e presents the influencing factors of depression in disabled elderly. Univariate logistic regression and multivariate logistic regression were employed to identify independent risk factors for depression among disabled elderly, resulting in the selection of five significant predictive factors for constructing a depression prediction model in disabled elderly, as shown in Tables\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e and \u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. \" poor self-rated health\", \"pain\", \" shorter sleep duration at night\", \"lack of caregivers\" and \"cognitive impairment\" emerged as independent risk factors in the logistic regression analysis.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eResults of univariate logistic regression analysis\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eB\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eS.E\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eWald\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003edf\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSig.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eExp(B)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003e95%C.I.for EXP(B)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.510\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.088\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e33.618\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.600\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.505\u0026ndash;0.713\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e8.370\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.015\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e60\u0026ndash;69 vs. \u0026gt;=80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.444\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.154\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e8.256\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.558\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.151\u0026ndash;2.109\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e70\u0026ndash;79 vs. \u0026gt;=80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.350\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.161\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4.753\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.029\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.419\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.036\u0026ndash;1.994\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eResidence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.413\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.106\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e15.294\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.512\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.229\u0026ndash;1.859\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducation level\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e25.552\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eilliteracy vs. high school and above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0. 978\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.229\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e18.244\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e2.660\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.698\u0026ndash;4.166\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eelementary school level vs. high school and above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.910\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.227\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e16.111\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e2.483\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.593\u0026ndash;3.872\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ejunior high school level vs. high school and above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.553\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.248\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4.971\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.026\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.738\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.069\u0026ndash;2.825\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHealth self-assessment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e162.176\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003egood vs. poor\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-1.534\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.154\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e98.787\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.216\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.159\u0026ndash;0.292\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ecommon vs. poor\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-1.004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.095\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e110.951\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.367\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.304\u0026ndash;0.442\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePain\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e178.705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003enone vs. often\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-1.669\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.127\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e172.755\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.188\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.147\u0026ndash;0.242\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOccasionally vs. often\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.942\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.106\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e79.041\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.390\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.317\u0026ndash;0.480\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSleep at night\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e86.648\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;6h vs. \u0026gt;8h\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.726\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.148\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e24.217\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e2.067\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.548\u0026ndash;2.761\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e6-8h vs. \u0026gt;8h\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.126\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.147\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.733\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.392\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.882\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.662\u0026ndash;1.176\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSocial activities\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.249\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.087\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e8.219\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.779\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.657\u0026ndash;0.924\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCaregiver\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.592\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.095\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e38.790\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.553\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.459\u0026ndash;0.666\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWork\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e75.693\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eno problem vs. unable\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.899\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.114\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e61.785\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.407\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.325\u0026ndash;0.509\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eshort time vs. unable\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.138\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.108\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.633\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.201\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.871\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.705\u0026ndash;1.076\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCognitive impairment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.291\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.088\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e11.046\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.338\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.127\u0026ndash;1.588\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDeposit\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.504\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.116\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e18.742\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.655\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.317\u0026ndash;2.079\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eResults of multivariate logistic regression analysis\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eB\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eS.E\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eWald\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003edf\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSig.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eExp(B)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003e95%C.I.for EXP(B)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHealth self-assessment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e53.612\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003egood vs. poor\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.985\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.171\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e33.025\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.373\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.267\u0026ndash;0.522\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ecommon vs. poor\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.671\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.107\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e39.278\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.511\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.414\u0026ndash;0.630\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePain\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e60.979\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003enone vs. often\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-1.088\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.141\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e59.651\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.337\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.256\u0026ndash;0.444\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOccasionally vs. often\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.580\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.114\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e25.746\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.560\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.447-0.700\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSleep at night\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e36.563\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;6h vs. \u0026gt;8h\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.497\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.162\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9.427\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.644\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.197\u0026ndash;2.257\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e6-8h vs. \u0026gt;8h\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.110\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.161\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.469\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.494\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.895\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.653\u0026ndash;1.228\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCaregiver\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.597\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.104\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e32.638\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.551\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.449\u0026ndash;0.676\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCognitive impairment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.222\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.101\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4.859\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.028\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.249\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.025\u0026ndash;1.522\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003ePredictive performance of disabled elderly\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e and Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e showed that except for decision trees, the use of LR and ML techniques for predicting depression among disabled elderly individuals is deemed acceptable. In the training set, XGBoost exhibited good predictive performance (AUC\u0026thinsp;=\u0026thinsp;0.76). In the validation set, traditional LR showed better predictive performance (AUC\u0026thinsp;=\u0026thinsp;0.73). Decision trees produced inconclusive results in both the training and validation sets (AUC\u0026thinsp;\u0026lt;\u0026thinsp;0.70); however, the other models did not show significant differences in AUC. The overall predictive performance, as proved by the Brier score, demonstrated favorable outcomes. Figure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e illustrated the calibration plot, showing good consistency between the predicted probabilities of LR and XGBoost and the actual observations. In Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, within the threshold probability range of 0.15 to 0.89 for depression among disabled elderly individuals, implementing a selective intervention strategy yielded higher net gains than the default approach of intervention or non-intervention with all patients.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eComparison of Predictive Model Performance\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"10\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAUC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003ePPV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNPV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eprecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003erecall\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eBrier-score\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTraining set\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.73\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.66\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.59\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.20\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.70\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.70\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.21\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.78\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.82\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.22\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.82\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.22\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eXGB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.76\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.78\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.71\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.71\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.20\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eValidation set\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.73\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.78\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.76\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.20\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.66\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.76\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.51\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.61\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.73\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.21\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.78\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.83\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.22\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.66\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.51\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.21\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eXGB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.71\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.59\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.73\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.21\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eVisualization by SHAP\u003c/h2\u003e \u003cp\u003eWe were using the SHAP algorithm to intuitively display the independent risk factors for predicting depression among functionally impaired older adults using the XGBOOST model. (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA, B) demonstrated the ranking of the importance of risk factors in descending order. Self-rated health has the most vital predictive ability, followed by pain and shorter sleep duration at night. (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC) SHAP provides feature interaction diagrams to identify features suitable for combination. We also provide two typical examples, one predicting no depression (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD) and the other predicting depression (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE), to demonstrate the interpretability of the model.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eConstruction of nomogram\u003c/h2\u003e \u003cp\u003eBased on the logistic regression analysis results, a nomogram is constructed for the independent risk factors mentioned above. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e, each axis represents a specific variable, and the corresponding values for each variable are found along the axis, typically marked with scales. Then, summing these values yields a total score corresponding to the predicted probability, indicating the higher the total score, the greater the likelihood of depression among disabled elderly individuals.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eWe investigated the contributing factors between disability among elderly individuals and depression using cross-sectional data from a representative Chinese population. The predictive factors consisted of \" health self-assessment\", \"pain\", \" sleep at night\", \"caregiver\" and \"cognitive impairment.\" This study was the first to predict the risk of depression in disabled individuals aged 60 years or older in China, and it also compared various machine learning techniques with traditional logistic regression.\u003c/p\u003e \u003cp\u003eAll models, except for decision trees, demonstrate acceptable discriminative capability; however, their performance falls short of the desired threshold (AUC\u0026thinsp;\u0026ge;\u0026thinsp;0.9) [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. This may be because the concept of depressive tendencies involves multiple aspects and diagnostic approaches, and currently, there are no valuable biomarkers or biological screening tests in clinical practice [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Conversely, prediction models have demonstrated good performance in diseases characterized by well-defined conditions, such as stroke and diabetes [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. It is worthwhile to consider some depression-related risk factors [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Therefore, it is of practical significance to predict depression in disabled elderly individuals by identifying crucial predictive factors [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. Risk prediction models for depression, integrating variables such as cognitive assessments and self-rated health, achieve satisfactory performance when applied to cross-sectional data.\u003c/p\u003e \u003cp\u003eThe sensitivity values of all models (ranging from 0.74 to 0.80) are higher compared to the specificity values (ranging from 0.51 to 0.58). Greater sensitivity contributes to a lower false-negative rate, facilitating the identification of a larger proportion of elderly individuals exhibiting depressive tendencies. This enhances the awareness of primary healthcare workers toward disabled older adults with depressive tendencies and allows for early prevention strategies targeting this specific population group. Furthermore, the results of DCA indicate that all models can be used in clinical practice within a reasonable range of threshold probabilities[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. Predictive models built using readily accessible variables will be employed in diverse scenarios for the identification of individuals at risk of depression and the implementation of preventive interventions[\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e\"Poor self-rated health,\" \"pain,\" \" shorter sleep duration at night,\" \" absence of caregivers,\" and \"cognitive impairment\" are high-risk factors for depression in disabled elderly individuals. The most prominent risk factor identified in this study is \"poor self-rated health,\" which measures self-perceived health. Previous research has confirmed the correlation between self-rated health and depression.[\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. For individuals with impaired functioning, perceiving oneself as physically unhealthy is more likely to trigger severe depressive symptoms than perceiving oneself as physically healthy. This finding implies that the elderly population's perception of health may exert a more substantial influence on their depressive symptoms.\u003c/p\u003e \u003cp\u003eDepression and pain share significant pathophysiological overlap, with a higher prevalence of depression observed in patients with chronic pain compared to those without pain[\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. Furthermore, the simultaneous presence of pain and depressive symptoms adversely affects individuals[\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. Over 60% of individuals with depression experience chronic pain symptoms[45]. The comorbidity between pain and depression holds particular significance in clinical settings, as the simultaneous presence of chronic pain and depression in older adults presents treatment challenges[\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. Furthermore, depression and pain have a bidirectional relationship where they can act as risk factors for each other[\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e47\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eShort sleep duration may increase daytime fatigue, which can lead to adverse events and emotions caused by fatigue, ultimately resulting in depression[\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e48\u003c/span\u003e].In addition, research has shown that longer sleep duration is associated with lower levels of physical activity, which is beneficial for reducing the risk of depression[\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e49\u003c/span\u003e]. Through elevation of neurotransmitter levels, specifically dopamine and serotonin, and enhancement of brain noradrenergic synaptic transmission[\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e50\u003c/span\u003e], this mechanism promotes endorphin secretion[\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e51\u003c/span\u003e], diminishes stress stimuli[\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e52\u003c/span\u003e], and enhances self-efficacy and self-esteem. A longitudinal study revealed a robust correlation between depression scores and sleep duration, indicating a substantial impact of inadequate sleep on depression[\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e53\u003c/span\u003e]. Furthermore, another prospective study showed that shorter sleep duration is associated with increased severity of depressive symptoms[\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e54\u003c/span\u003e]. There is a strong bidirectional relationship between sleep and depression, where reduced sleep duration serves as the strongest predictor for increased acute depressive symptoms. Moreover, more than 80% of individuals with depression experience disturbances in their sleep patterns.\u003c/p\u003e \u003cp\u003eInadequate social support has a detrimental impact on mental health[\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e55\u003c/span\u003e], whereas sufficient social support exerts a positive influence on the well-being of individuals experiencing depression. Previous studies have consistently identified spousal care for elderly individuals as a protective factor against depression. This is especially true for males, as poor marital relationships or the absence of a partner at home are related to depression in elderly individuals[\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e56\u003c/span\u003e]. Caregivers can provide accommodation and meals, thereby reducing the need for hospitalization. They can also make more use of social networks and provide support by accompanying individuals during treatment. Insightful family members or close friends can serve as an \"early warning system,\" enabling early intervention for individuals who may be at risk of depression[\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e57\u003c/span\u003e]. Compared to elderly individuals with caregivers, empty nesters show a significantly higher tendency towards depression than non-empty nesters[\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e58\u003c/span\u003e]. Studies have demonstrated a prevalence rate exceeding 70% for depression or depressive symptoms among empty nesters in China[\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e59\u003c/span\u003e, \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e60\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eDepression may co-occur with or even precede dementia, which is characterized by diffuse cognitive impairment[\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e61\u003c/span\u003e]. Studies have indicated an association between cognitive dysfunction and late-life depression, as well as adverse reactions to antidepressant medications[\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e62\u003c/span\u003e]. Furthermore, cognitive impairment plays a significant role in the development of depression[\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e63\u003c/span\u003e]. Cognitive impairments are observed in individuals at the onset of depression. Recurrent episodes of depression exhibit more significant cognitive impairments compared to single episodes, particularly in processing speed, executive function, language learning, and memory[\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e64\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWe have developed a risk prediction model for depression among disabled elderly individuals. By incorporating five high-risk factors: \" health self-assessment\", \"pain\", \" sleep at night\", \"caregiver\" and \"cognitive impairment.\" We have developed a web-based clinical support system that is user-friendly and community-oriented. This system will facilitate the early identification of elderly individuals with depressive tendencies by themselves and caregivers, promoting proactive care and enhancing healthcare allocation through targeted interventions for disease prevention and management.\u003c/p\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eStrengths and limitations\u003c/h2\u003e \u003cp\u003eTo our knowledge, this is the first nomogram constructed based on the Chinese elderly population to predict depression among disabled older adults. This nomogram accurately identifies individuals with high-risks by incorporating selected independent risk factors. This tool will assist caregivers and healthcare practitioners in implementing timely intervention strategies to prevent depression among disabled older adults.\u003c/p\u003e \u003cp\u003eInevitably, this study has some limitations. Firstly, caution should be exercised when extrapolating the findings of this study to other countries as it is solely based on the Chinese population. Secondly, the presence of biases induced by missing data must be acknowledged. To ensure the robustness of the predictive model for depression among disabled elderly individuals, participants with missing data or abnormal values were excluded from the analysis. Thirdly, we could not obtain other important predictive factors, such as specific diets and physical activity, that were not collected in the CHARLS dataset[\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e65\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn the study, the logistic regression and XGBoost models demonstrated good discrimination, calibration, overall predictive performance, and clinical utility in predicting depression among disabled elderly individuals. A straightforward and efficient preliminary clinical support system was developed based on the logistic regression model, showing promise to significantly reduce the burden on users and help healthcare service providers manage depression.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"35.2803738317757%\" valign=\"top\"\u003e\n \u003cp\u003eADL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"64.7196261682243%\" valign=\"top\"\u003e\n \u003cp\u003eactivity of daily living\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"35.2803738317757%\" valign=\"top\"\u003e\n \u003cp\u003eIADL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"64.7196261682243%\" valign=\"top\"\u003e\n \u003cp\u003einstrumental activity of daily living\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"35.2803738317757%\" valign=\"top\"\u003e\n \u003cp\u003ePSMS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"64.7196261682243%\" valign=\"top\"\u003e\n \u003cp\u003ephysical self-maintenance scale\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"35.2803738317757%\" valign=\"top\"\u003e\n \u003cp\u003eCESD-10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"64.7196261682243%\" valign=\"top\"\u003e\n \u003cp\u003e10-item Center for Epidemiologic Studies Depression Scale\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"35.2803738317757%\" valign=\"top\"\u003e\n \u003cp\u003eROC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"64.7196261682243%\" valign=\"top\"\u003e\n \u003cp\u003ereceiver operating characteristic\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"35.2803738317757%\" valign=\"top\"\u003e\n \u003cp\u003eAUC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"64.7196261682243%\" valign=\"top\"\u003e\n \u003cp\u003earea under the curve\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"35.2803738317757%\" valign=\"top\"\u003e\n \u003cp\u003eDCA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"64.7196261682243%\" valign=\"top\"\u003e\n \u003cp\u003edecision curve analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"35.2803738317757%\" valign=\"top\"\u003e\n \u003cp\u003eSHAP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"64.7196261682243%\" valign=\"top\"\u003e\n \u003cp\u003eShapley Additive exPlanations\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"35.2803738317757%\" valign=\"top\"\u003e\n \u003cp\u003eXGBoost\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"64.7196261682243%\" valign=\"top\"\u003e\n \u003cp\u003eeXtreme Gradient Boosting\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"35.2803738317757%\" valign=\"top\"\u003e\n \u003cp\u003eHDL-C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"64.7196261682243%\" valign=\"top\"\u003e\n \u003cp\u003ehigh-density lipoprotein cholesterol\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"35.2803738317757%\" valign=\"top\"\u003e\n \u003cp\u003eLR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"64.7196261682243%\" valign=\"top\"\u003e\n \u003cp\u003elogistic regression\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"35.2803738317757%\" valign=\"top\"\u003e\n \u003cp\u003eML\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"64.7196261682243%\" valign=\"top\"\u003e\n \u003cp\u003emachine learning\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"35.2803738317757%\" valign=\"top\"\u003e\n \u003cp\u003eSVM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"64.7196261682243%\" valign=\"top\"\u003e\n \u003cp\u003esupport vector machine\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank the Chinese Health and Retirement Longitudinal Study team for generously providing the required data. Additionally, our appreciation goes to the participating students, volunteers, and staff who contributed to this research endeavor.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eShanshan Hong conceptualized and planned the overall study, conducted data analysis, and drafted the manuscript. Bingqian Lu,Yan Jiang and Shaobing Wang critically reviewed the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFundings\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFaculty development grants from Hubei University of Medicine (2020QDJZR016)\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWuhan Optoelectronics National Research Center 2021 Open Fund Project\u0026nbsp;(2021WNLOKF020).\u003c/p\u003e\n\u003cp\u003eThe funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe database was used from the China Health and Retirement Longitudinal Study. (http://charls.pku.edu.cn/).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was conducted following the principles of the Helsinki Declaration and was approved by the Biomedical Ethics Committee of Peking University. All participants signed informed consent forms before their participation, which were approved by the Ethics Review Committee of Peking University (IRB00001052-11015).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eZheng X, Pang L, Chen G, Huang C, Liu L, Zhang L. Challenge of Population Aging on Health. Public Health Challenges in Contemporary Ch-ina: An Interdisciplinary Perspective.2016:43\u0026ndash;53.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-3-662-47753-3_4\u003c/span\u003e\u003cspan address=\"10.1007/978-3-662-47753-3_4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStordal E, Mykletun A, Dahl AA. The association between age and depression in the general population: a multivariate examination. Acta psychiatrica Scandinavica. 2003;107:132\u0026ndash;41.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWhiteford HA, Ferrari AJ, Degenhardt L, Feigin V, Vos T. The Global Burden of Mental, Neurological and Substance Use Disorders: An Analysis from the Global Burden of Disease Study 2010. PLoS ONE. 2015;10:14.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMueller TI, Kohn R, Leventhal N, Leon AC, Solomon D, Coryell W, Endicott J, Alexopoulos GS, Keller MB. The course of depression in elderly patients. Am J Geriatr Psychiatry. 2004;12:22\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMitchell AJ, Subramaniam H. Prognosis of depression in old age compared to middle age: a systematic review of comparative studies. Am J Psychiatry. 2005;162:1588\u0026ndash;601.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVerhaak PF, Dekker JH, de Waal MW, van Marwijk HW. Comijs HC. Depression, disability and somatic diseases among elderly. J Affect Disord. 2014;167:187\u0026ndash;91.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRobinson-Whelen S, Taylor HB, Hughes RB, Wenzel L, Nosek MA. Depression and depression treatment in women with spinal cord injury. Top Spinal Cord Inj Rehabil. 2014;20:23\u0026ndash;31.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWanjau MN, M\u0026ouml;ller H, Haigh F, Milat A, Hayek R, Lucas P, Veerman JL. Physical Activity and Depression and Anxiety Disorders in Australia: A Lifetable Analysis. AJPM Focus. 2023;2:100030.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNoh JW, Kwon YD, Park J, Oh IH, Kim J. Relationship between Physical Disability and Depression by Gender: A Panel Regression Model. PLoS ONE. 2016;11:e0166238.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDjernes JK. Prevalence and predictors of depression in populations of elderly: a review. Acta Psychiatr Scand. 2006;113:372\u0026ndash;87.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHassler AP, Menasalvas E, Garc\u0026iacute;a-Garc\u0026iacute;a FJ, Rodr\u0026iacute;guez-Ma\u0026ntilde;as L, Holzinger A. Importance of medical data preprocessing in predictive modeling and risk factor discovery for the frailty syndrome. BMC Med Inform Decis Mak. 2019;19:33.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAkbari G, Nikkhoo M, Wang L, Chen CPC, Han DS, Lin YH, Chen HB, Cheng CH. Frailty Level Classification of the Community Elderly Using Microsoft Kinect-Based Skeleton Pose: A Machine Learning Approach. Sens (Basel). 2021;21. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/s21124017\u003c/span\u003e\u003cspan address=\"10.3390/s21124017\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBakhtiari A, Hashemi M, Hosseini SR, Omidvar S, Bijani A, Khairkhah F. The Relationship between Depression and Metabolic Syndrome in the Elderly Population: The Cohort Aging Study. Iran J Psychiatry. 2018;13:230\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTyler Williamson P, Aponte-Hao S, Mele B, Lethebe BC, Leduc C, Thandi M, Katz A, Wong ST. Developing and Validating a Primary Care EMR-based Frailty Definition using Machine Learning. Int J Popul Data Sci. 2020;5:1344.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePark C, Mishra R, Sharafkhaneh A, Bryant MS, Nguyen C, Torres I, Naik AD, Najafi B. Digital Biomarker Representing Frailty Phenotypes: The Use of Machine Learning and Sensor-Based Sit-to-Stand Test. Sens (Basel). 2021;21. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/s21093258\u003c/span\u003e\u003cspan address=\"10.3390/s21093258\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShin C, Park MH, Lee SH, Ko YH, Kim YK, Han KM, Jeong HG, Han C. Usefulness of the 15-item geriatric depression scale (GDS-15) for classifying minor and major depressive disorders among community-dwelling elders. J Affect Disord. 2019;259:370\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao Y, Hu Y, Smith JP, Strauss J, Yang G. Cohort profile: the China Health and Retirement Longitudinal Study (CHARLS). Int J Epidemiol. 2014;43:61\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKatz S, Ford AB, Moskowitz RW, Jackson BA, Jaffe MW, STUDIES OF ILLNESS IN THE AGED. THE INDEX OF ADL: A STANDARDIZED MEASURE OF BIOLOGICAL AND PSYCHOSOCIAL FUNCTION. JAMA. 1963;185:914\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCovinsky KE, Hilton J, Lindquist K, Dudley RA. Development and validation of an index to predict activity of daily living dependence in community-dwelling elders. Med Care. 2006;44:149\u0026ndash;57.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKingston A, Comas-Herrera A, Jagger C. Forecasting the care needs of the older population in England over the next 20 years: estimates from the Population Ageing and Care Simulation (PACSim) modelling study. Lancet Public Health. 2018;3:e447\u0026ndash;55.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou L, Ma X, Wang W. Relationship between Cognitive Performance and Depressive Symptoms in Chinese Older Adults: The China Health and Retirement Longitudinal Study (CHARLS). J Affect Disord. 2021;281:454\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDong L, Freedman VA, Mendes de Leon CF. The Association of Comorbid Depression and Anxiety Symptoms With Disability Onset in Older Adults. Psychosom Med. 2020;82:158\u0026ndash;64.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen CM, Liu LF. The effect of disability and depression on cognitive function and screening factors. Arch Gerontol Geriatr. 2017;73:154\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGortmaker SL, Hosmer DW, Lemeshow SJCS. Applied Logistic Regression. 3rd ed. Hoboken: John Wiley \u0026amp; Sons; 2013.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang H, Chen D, Shao J, Zou P, Cui N, Tang L, Wang X, Wang D, Ye Z. External Validation of the Prognostic Prediction Model for 4-Year Risk of Metabolic Syndrome in Adults: A Retrospective Cohort Study. Diabetes Metab Syndr Obes. 2021;14:3027\u0026ndash;34.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang YQ, Liang CH, He L, Tian J, Liang CS, Chen X, Ma ZL, Liu ZY. Development and Validation of a Radiomics Nomogram for Preoperative Prediction of Lymph Node Metastasis in Colorectal Cancer. J Clin Oncol. 2016;34:2157\u0026ndash;64.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSteyerberg EW, Vickers AJ, Cook NR, Gerds T, Gonen M, Obuchowski N, Pencina MJ, Kattan MW. Assessing the performance of prediction models: a framework for traditional and novel measures. Epidemiology.2010;21:128\u0026thinsp;\u0026ndash;\u0026thinsp;38.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKerr KF, Brown MD, Zhu K, Janes H. Assessing the Clinical Impact of Risk Prediction Models With Decision Curves: Guidance for Correct Interpretation and Appropriate Use. J Clin Oncol. 2016;34:2534\u0026ndash;40.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLundberg S, Lee S-I. A unifed approach to interpreting model predictions.ArXiv170507874 Cs Stat. 2017 [cited 1 Sep 2020]. Available: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://arxiv.org/abs/1705.07874\u003c/span\u003e\u003cspan address=\"http://arxiv.org/abs/1705.07874\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAthanasiou M, Sfrintzeri K, Zarkogianni K, Thanopoulou AC, Nikita KS. An explainable XGBoost\u0026ndash;based approach towards assessing the risk of cardiovascular disease in patients with Type 2 Diabetes Mellitus. 2020 IEEE 20th International Conference on Bioinformatics and Bioengineering (BIBE).2020;859\u0026ndash;864.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJoseph AJapa. Parametric inference with universal function approximators. arXiv preprint arXiv: 1903.04209 ML Stat. 2019 [cited 4 Oct 2020]. Available: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.48550/arXiv.1903.04209\u003c/span\u003e\u003cspan address=\"10.48550/arXiv.1903.04209\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLundberg SM, Nair B, Vavilala MS, Horibe M, Eisses MJ, Adams T, Liston DE, Low DK-W. Newman S-F and Kim JJNbe. Explainable machine-learning predictions for the prevention of hypoxaemia during surgery. 2018;2:749\u0026ndash;60.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang QQ, Yu SC, Qi X, Hu YH, Zheng WJ, Shi JX, Yao HY. [Overview of logistic regression model analysis and application]. Zhonghua Yu Fang Yi Xue Za Zhi. 2019;53:955\u0026ndash;60.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcCarron RM, Shapiro B, Rawles J, Luo. J Depress Ann Intern Med. 2021;174:Itc65\u0026ndash;itc80.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePark E, Chang HJ, Nam HS. A Bayesian Network Model for Predicting Post-stroke Outcomes With Available Risk Factors. Front Neurol. 2018;9:699.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTing DSW, Cheung CY, Lim G, Tan GSW, Quang ND, Gan A, Hamzah H, Garcia-Franco R, San Yeo IY, Lee SY, Wong EYM, Sabanayagam C, Baskaran M, Ibrahim F, Tan NC, Finkelstein EA, Lamoureux EL, Wong IY, Bressler NM, Sivaprasad S, Varma R, Jonas JB, He MG, Cheng CY, Cheung GCM, Aung T, Hsu W, Lee ML, Wong TY. Development and Validation of a Deep Learning System for Diabetic Retinopathy and Related Eye Diseases Using Retinal Images From Multiethnic Populations With Diabetes. JAMA. 2017;318:2211\u0026ndash;23.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrown CS. Depression and anxiety disorders. Obstet Gynecol Clin North Am. 2001;28:241\u0026ndash;68.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFried EI, Epskamp S, Nesse RM, Tuerlinckx F, Borsboom D. What are 'good' depression symptoms? Comparing the centrality of DSM and non-DSM symptoms of depression in a network analysis. J Affect Disord. 2016;189:314\u0026ndash;20.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTarekegn A, Ricceri F, Costa G, Ferracin E, Giacobini M. Predictive Modeling for Frailty Conditions in Elderly People: Machine Learning Approaches. JMIR Med Inform. 2020;8:e16678.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDong BR, Gu XQ, Chen HY, Gu J, Pan ZG. Development and Validation of a Nomogram to Predict Frailty Progression in Nonfrail Chinese Community-Living Older Adults. J Am Med Dir Assoc. 2021;22:2571\u0026ndash;2578e4.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJahn DR, Cukrowicz KC. Self-rated health as a moderator of the relation between functional impairment and depressive symptoms in older adults. Aging Ment Health. 2012;16:281\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLyness JM, King DA, Conwell Y, Duberstein PR, Eberly S, S\u0026ouml;rensen SM, Caine ED. Self-rated health, depression, and one-year health outcomes in older primary care patients. Am J Geriatr Psychiatry. 2004;12:110\u0026ndash;3.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDepression and pain. J Clin Psychiatry. 2008;69:1970\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRobinson MJ, Edwards SE, Iyengar S, Bymaster F, Clark M, Katon W. Depression and pain. Front Biosci (Landmark Ed).2009;14:5031-51.Chronic pain and major depressive disorder in the general population. J Psychiatr Res. 2010;44:454\u0026ndash;61.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFishbain DA, Cutler R, Rosomoff HL, Rosomoff RS. Chronic pain-associated depression: antecedent or consequence of chronic pain? A review. Clin J Pain. 1997;13:116\u0026ndash;37.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZis P, Daskalaki A, Bountouni I, Sykioti P, Varrassi G, Paladini A. Depression and chronic pain in the elderly: links and management challenges. Clin Interv Aging. 2017;12:709\u0026ndash;20.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhai L, Zhang H, Zhang D, SLEEP DURATION AND, DEPRESSION. AMONG ADULTS: A META-ANALYSIS OF PROSPECTIVE STUDIES. Depress Anxiety. 2015;32:664\u0026ndash;70.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStranges S, Dorn JM, Shipley MJ, Kandala NB, Trevisan M, Miller MA, Donahue RP, Hovey KM, Ferrie JE, Marmot MG, Cappuccio FP. Correlates of short and long sleep duration: a cross-cultural comparison between the United Kingdom and the United States: the Whitehall II Study and the Western New York Health Study. Am J Epidemiol. 2008;168:1353\u0026ndash;64.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWeicker H, Str\u0026uuml;der HK. Influence of exercise on serotonergic neuromodulation in the brain. Amino Acids. 2001;20:35\u0026ndash;47.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJanal MN, Colt EWD, Clark CW, Glusman M. Pain sensitivity, mood and plasma endocrine levels in man following long-distance running: effects of naloxone. Pain. 1984;19:13\u0026ndash;25.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSalmon P. Effects of physical exercise on anxiety, depression, and sensitivity to stress: a unifying theory. Clin Psychol Rev. 2001;21:33\u0026ndash;61.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFirth-Cozens J. Individual and organizational predictors of depression in general practitioners. Br J Gen Pract. 1998;48:1647\u0026ndash;51.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBuckley T, Bartrop R, McKinley S, Ward C, Bramwell M, Roche D, Mihailidou AS, Morel-Kopp MC, Spinaze M, Hocking B, Goldston K, Tennant C, Tofler G. Prospective study of early bereavement on psychological and behavioural cardiac risk factors. Intern Med J. 2009;39:370\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrown GW, Andrews B. Social support and depression. Dynamics of stress: Physiological, psychological and social perspectives.1986;257\u0026ndash;282.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSonnenberg CM, Deeg DJ, van Tilburg TG, Vink D, Stek ML, Beekman AT. Gender differences in the relation between depression and social support in later life. Int Psychogeriatr. 2013;25:61\u0026ndash;70.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSmith L, Hill N, Kokanovic R. Experiences of depression, the role of social support and its impact on health outcomes. J Ment Health. 2015;24:342\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhai Y, Yi H, Shen W, Xiao Y, Fan H, He F, Li F, Wang X, Shang X, Lin J. Association of empty nest with depressive symptom in a Chinese elderly population: A cross-sectional study. J Affect Disord. 2015;187:218\u0026ndash;23.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXie LQ, Zhang JP, Peng F, Jiao NN. Prevalence and related influencing factors of depressive symptoms for empty-nest elderly living in the rural area of YongZhou, China. Arch Gerontol Geriatr. 2010;50:24\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSu D, Wu XN, Zhang YX, Li HP, Wang WL, Zhang JP, Zhou LS. Depression and social support between China' rural and urban empty-nest elderly. Arch Gerontol Geriatr. 2012;55:564\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKiosses DN, Alexopoulos GS. IADL functions, cognitive deficits, and severity of depression: a preliminary study. Am J Geriatr Psychiatry. 2005;13:244\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorimoto SS, Kanellopoulos D, Manning KJ, Alexopoulos GS. Diagnosis and treatment of depression and cognitive impairment in late life. Ann N Y Acad Sci. 2015;1345:36\u0026ndash;46.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrahek I, Shenhav A, Musslick S, Krebs RM, Koster EHW. Motivation and cognitive control in depression. Neurosci Biobehav Rev. 2019;102:371\u0026ndash;81.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVarghese S, Frey BN, Schneider MA, Kapczinski F, de Azevedo Cardoso T. Functional and cognitive impairment in the first episode of depression: A systematic review. Acta Psychiatr Scand. 2022;145:156\u0026ndash;85.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeng LN, Hsiao FY, Lee WJ, Huang ST, Chen LK. Comparisons Between Hypothesis- and Data-Driven Approaches for Multimorbidity Frailty Index: A Machine Learning Approach. J Med Internet Res. 2020;22:e16213.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-psychiatry","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bpsy","sideBox":"Learn more about [BMC Psychiatry](http://bmcpsychiatry.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bpsy/default.aspx","title":"BMC Psychiatry","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Disability, Depression, Predictive Model, Elderly individuals","lastPublishedDoi":"10.21203/rs.3.rs-3874875/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3874875/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eGiven the accelerated aging population in China, the number of disabled elderly individuals is increasing, depression has been a common mental disorder among older adults. This study aims to establish an effective model for predicting depression risks among disabled elderly individuals.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eThe data for this study was obtained from the 2018 China Health and Retirement Longitudinal Study (CHARLS). In this study, disability was defined as a functional impairment in at least one activity of daily living (ADL) or instrumental activity of daily living (IADL). Depressive symptoms were assessed by using the 10-item Center for Epidemiologic Studies Depression Scale (CES-D10). We employed SPSS 27.0 to select independent risk factor variables associated with depression among disabled elderly individuals. Subsequently, a predictive model for depression in this population was constructed using R 4.3.0. The model's discrimination, calibration, and clinical net benefits were assessed using receiver operating characteristic (ROC) curves, calibration plots, and decision curves.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eIn this study, a total of 3,107 elderly individuals aged\u0026thinsp;\u0026ge;\u0026thinsp;60 years with disabilities were included. Poor self-rated health, pain, absence of caregivers, cognitive impairment, and shorter sleep duration were identified as independent risk factors for depression in disabled elderly individuals. The XGBoost model demonstrated better predictive performance in the training set, while the logistic regression model showed better predictive performance in the validation set, with AUC of 0.76 and 0.73, respectively. The calibration curve and Brier score (Brier: 0.20) indicated a good model fit. Moreover, decision curve analysis confirmed the clinical utility of the model.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eThe predictive model exhibits outstanding predictive efficacy, greatly assisting healthcare professionals and family members in evaluating depression risks among disabled elderly individuals. Consequently, it enables the early identification of elderly individuals at high risks for depression.\u003c/p\u003e","manuscriptTitle":"Comparison of logistic regression and machine learning methods for predicting depression risks among disabled elderly individuals: results from the China Health and Retirement Longitudinal Study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-01-24 18:57:18","doi":"10.21203/rs.3.rs-3874875/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-07-20T14:42:21+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-07-06T18:37:06+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-07-01T01:13:38+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"307661406561648944060618870719524836127","date":"2024-06-30T07:46:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"77403697344482400812308685797994685486","date":"2024-06-27T01:58:10+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"106200339482150063752877562818290682076","date":"2024-06-26T11:15:13+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"24146369684552356338703357009329608312","date":"2024-05-25T12:06:48+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-05-02T16:38:37+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-04-30T19:34:23+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2024-01-22T07:39:05+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-01-22T07:33:01+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Psychiatry","date":"2024-01-18T06:19:52+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-psychiatry","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bpsy","sideBox":"Learn more about [BMC Psychiatry](http://bmcpsychiatry.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bpsy/default.aspx","title":"BMC Psychiatry","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"ac349f2d-5558-4f78-8c5f-d81c32ff8848","owner":[],"postedDate":"January 24th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-02-17T16:03:34+00:00","versionOfRecord":{"articleIdentity":"rs-3874875","link":"https://doi.org/10.1186/s12888-025-06577-x","journal":{"identity":"bmc-psychiatry","isVorOnly":false,"title":"BMC Psychiatry"},"publishedOn":"2025-02-14 15:57:53","publishedOnDateReadable":"February 14th, 2025"},"versionCreatedAt":"2024-01-24 18:57:18","video":"","vorDoi":"10.1186/s12888-025-06577-x","vorDoiUrl":"https://doi.org/10.1186/s12888-025-06577-x","workflowStages":[]},"version":"v1","identity":"rs-3874875","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3874875","identity":"rs-3874875","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}