A relatively accurate prediction model for the risk of developing mild cognitive impairment in patients with sarcopenia: Evidence from the CHARLS

doi:10.21203/rs.3.rs-4932323/v1

Download PDF

Research Article

A relatively accurate prediction model for the risk of developing mild cognitive impairment in patients with sarcopenia: Evidence from the CHARLS

https://doi.org/10.21203/rs.3.rs-4932323/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

Sarcopenia significantly raises the risk of cognitive impairments in older adults. Early warning of mild cognitive impairment (MCI) in those with sarcopenia is crucial for timely intervention.

Aims

To construct an accurate prediction model for screening MCI in sarcopenia population.

Methods

We combined machine learning and deep learning techniques to analyze data from 597 sarcopenia patients in the China Health and Retirement Longitudinal Study (CHARLS). Our model predicts MCI incidence over the next four years, categorizing patients into low, intermediate, and high-risk groups based on their risk levels.

Results

The model was constructed using CHARLS data from 2011–2015 and included seven validated variables. It performed superior to logistic regression, achieving an Area Under the Curve (AUC) of 0.808 (95% CI: 0.704–0.899) for the test set and accurately classifying patients' risk in the training set. The deep learning model demonstrated a low false-positive rate of 1.63% for MCI in higher-risk groups. Independent validation using 2015–2018 CHARLS data confirmed the model’s efficacy, with an AUC of 0.76 (95% CI: 0.67–0.83). A convenient online tool to implement the model was developed at http://47.115.214.16:5000/.

Conclusions

This deep learning model effectively predicts MCI risk in sarcopenia patients, supporting early interventions. Its accuracy helps identify high-risk individuals, potentially improving patient care.

mild cognitive impairment

sarcopenia

CHARLS

prediction model

morbidity probability

online tool

The progression of cognitive function in humans is from normal age-related cognitive decline to MCI and then transition to Alzheimer's disease [1]. MCI is defined as a subjective and objective decline in functioning in one or more cognitive dimensions, such as language, memory, calculation, and orientation, compared with the past, without difficulty in activities of daily living [2]. However, it is characterized by an insidious and undetectable onset, rapid progression, easy evolution to dementia, and no effective drug treatment. According to the clinical practice guideline on MCI issued by the American Academy of Neurology, 14.4%-55.6% of patients with MCI may regain neurological integrity [3], thereby reducing the occurrence of dementia. Therefore, accurate and rapid identification of significant cognitive decline and the onset of MCI is essential for the prevention of dementia. However, the factors that cause cognitive decline are numerous and complex, and we cannot accurately predict the extent of cognitive decline, the risk of developing MCI, and etiologic prevention for those at high risk in all populations.

In patients with sarcopenia, the prevalence of MCI is 24.2% [4], and studies have shown that sarcopenia is 1.72 times more likely to develop MCI than non-sarcopenia and that patients with MCI have an annual progression rate of dementia of 10–20% [5]. We will focus our predictions on the sarcopenia population, which has a high prevalence of MCI and which causes severe adverse outcomes. Defined as a progressive and widespread accelerated loss of muscle mass and muscle function [4], sarcopenia is now formally recognized as a muscle disease [6], with an increasing prevalence in the elderly population every year. It is now considered an independent risk factor for MCI.

There is a common pathway, physiological, and pathological mechanisms between sarcopenia and MCI. Common metabolic (chronic inflammation, oxidative stress [7]), behavioral (exercise [8], socialization), and psychological (depressive symptoms [9]) factors that can affect both the body and cognition. The association between the two also has been studied. First, there is evidence that age-related loss of muscle mass, muscle function, and muscle strength occurs before cognitive decline. Thus, sarcopenia and its components can predict future cognitive impairment [10]. Second, numerous prior longitudinal studies have also shown that the components of sarcopenia, primarily low grip strength [11, 12] and slow gait speed [13], are significantly associated with the onset of MCI and can predict cognitive decline. Finally, the cognitive decline caused by sarcopenia is, to some extent, thought to be unrelated to the nervous system. Muscles may be able to influence cognition, such as various hormone-like proteins secreted by the muscles may reach the cognitive brain regions through the somatic circulation to directly affect cognition [14]. Therefore, perhaps there is a lack of direct mechanistic studies to prove the association between sarcopenia and MCI. Still, it is feasible to predict the risk of MCI using simple physical measures, which can help clinicians make decisions or guide interventional treatments.

We have not retrieved the prediction of MCI using sarcopenia-related characteristics and clinical and other metrics, so to address this issue, we developed a risk prediction model for MCI in sarcopenia using the CHARLS cohort. Increased computational power and the availability of big data have led to the application of DL to healthcare [15], which distinguishes itself from simple regression by allowing the analysis of nonlinear effects between characteristic variables and outcomes. In this study, we utilized the DL model to predict risk outcomes for MCI in patients with sarcopenia.

CHARLS is organized by the project team of the China Social Science Survey Center at Peking University. Its aim is to collect a set of high-quality microdata representative of China's middle-aged and elderly households and individuals aged 45 years and older and to promote interdisciplinary research on the issue of aging. The CHARLS national baseline survey was carried out in 2011 and is tracked every two to three years, covering 150 counties, 450 villages, and about 17,000 people in 10,000 households. To date, CHARLS has released five national surveys.

Participants

We have applied and used data from the four previous national surveys (2011, 2013, 2015, 2018) to collate physical measures of patients with sarcopenia. We retained only longitudinal data with normal cognitive function at baseline measurement and with indicators of cognitive outcomes, excluding those with severe physical impairments and psychiatric disorders. A total of 597 patients with sarcopenia were included in our subsequent analysis. The detailed information about CHARLS had been published in previous literature [16]. All CHARLS datasets can be downloaded at the CHARLS home page at http://charls.pku.edu.cn/en. The CHARLS survey project was approved by the Biomedical Ethics Committee of Peking University, and all participants were required to sign informed consent. Figure 1 shows the sample screening flowchart. Detailed information on data screening is provided in the supplementary material.

Assessment of sarcopenia

According to the Asian Working Group for Sarcopenia (AWGS 2019), sarcopenia is defined as loss of muscle mass, as well as low muscle strength and (or) poor physical performance [17]. We use Appendicular Skeletal Muscle (ASM) to represent muscle mass. ASM is generally measured using the Dual-energy X-ray Absorptiometry (DXA) or Bioelectrical Impedance Analysis (BIA). Chinese ASM can be evaluated using this formula: ASM = 0.193 × weight (kg) + 0.107 × height (cm) − 4.157 × gender − 0.037 × age (years) − 2.631, which is set to 1 if gender is male and 2 for female [18]. It was shown that the ASM calculated using this formula agrees well with DXA [19]. Low muscle mass was judged based on the 20% of the study population with the lowest gender specificity for height-adjusted muscle mass (SMI = ASM/height2) [4, 20]. Muscle strength was assessed by averaging two maximal grip strengths in both hands. The AWGS defined low muscle power as < 28 kg for men and < 18 kg for women [17]. Physical performance was assessed by walking speed, five-chair stand tests, and the Short Physical Performance Battery (SPPB), where SPPB consists of three tests in standing balance in addition to the first two tests, with 4 points each for a total of 12 points [21]. According to the AWGS 2019 recommendations, low physical performance was defined as gait speed < 1.0 m/s, five-chair stand tests ≥ 12s, or an SPPB score < 9 [17].

Assessment of MCI

Cognitive function was measured according to the methodology used in the American Health and Retirement Study (HRS) [22], where participants underwent a face-to-face assessment of four dimensions of cognitive functioning, namely orientation (5 points), memory (10 points), numeracy (5 points), and drawing ability (1 point) [23], for a total score of 21 points. We used Aging-Associated Cognitive Decline (AACD) to define MCI as being at least 1 standard deviation (SD) below the age norm [24]. All participants were grouped every 5 years of age, and participants in each age group who met the AACD criteria would be categorized as having MCI [4].

Covariates

We also considered demographic factors, health-related factors, and follow-up time that influenced MCI. Demographic characteristics included gender, age, place of residence (urban/rural), education level (primary and below primary school/middle school/high school or vocational school/college degree or higher.), and marital status (partnered/unpartnered) [25]. Health-related factors included history of smoking (yes/no), history of alcohol consumption (yes/no), daily sleep duration (in hours), depressive symptoms (yes/no), history of falls (yes/no), body mass index (BMI), and twelve chronic diseases associated with MCI [26] (hypertension, dyslipidemia, hyperglycemia, chronic lung disease, heart disease, liver disease, kidney disease, digestive system disease, malignant tumor, stroke, arthritis rheumatism, asthma). Among them, BMI was defined as weight (in kg) divided by the square of height (in m). Depression was assessed using the 10-item Center for Epidemiologic Studies Depression Scale (CESD-10), with a total score of 30, with a score of more than 10 suggesting the presence of depressive symptoms [16, 27].

Statistical analysis

Continuous variables were expressed as means ± standard deviations. Categorical variables were expressed as frequency (n) and proportion (%). A T-test was used to compare the measurement data between the two groups. Multi-group categorical data were compared using the rank sum test. We use Stata software to merge and filter the data. Data were filled with the R software missforest package. A feed-forward neural network model was constructed using Pytorch. The Receiver Operating Characteristic (ROC) curve represented the model's discrimination and was assessed by the AUC value. A two-sided P value less than or equal to 0.05 was considered statistically significant. Statistical analyses were performed using Stata17, R.4.2.2, and Pytorch 2.2.1.

Demographic characteristics of participants

597 patients with sarcopenia met the inclusion criteria, of which 137 had MCI. Table 1 provides the demographic characteristics of the participants. The average age of the sample population was 69.40.

Table 1

Demographic characteristics of the participants
Variables	Total(n = 597)
Age	69.40 ± 7.52
Gender
male	295(49.4)
female	302(50.6)
Residence
rural	281(47.1)
urban	316(52.9)
Companion
yes	449(75.2)
no	148(24.8)
Education
primary school and below	540(90.5)
middle school	34(5.7)
High school or vocational school	20(3.4)
College degree or above	3(0.5)

Selection of predictors

The 430 patients with sarcopenia from CHARLS (2011–2015) were used as the training and test sets for selecting predictors and model construction, focusing on respondents' general information, health status, physical measurements, and cognitive status. Considering 30 drivers of MCI, all had 80% or more completeness of the data, and the expression levels of all variables were standardized. We use ML for feature double filtering. The random forest model was constructed using the R software random forest package to rank the importance of features and exclude those with mean decrease Gini (MDG) of less than 1. The remaining 17 features were subjected to recursive feature elimination (re) using the caret package, and in the process of elimination, ten-fold cross-validation was used to avoid overfitting. After the dual method validation, the final 7 features entered the next modeling stage, and these metrics included SMI, ASM, BMI, SPPB, Depression, Diabetes, and Residence (Table 2). Among them, muscle mass index played a crucial role in the model's prediction, and ASM (P = 0.022) and SMI (P = 0.004) showed a significant difference between the MCI and non-MCI groups.

Table 2

Comparison of the characteristics of the population with and without MCI and ranking of the importance of the characteristics to the model
	Total(n = 430)	No(n = 329)	Yes(n = 101)	P-value	MDG	Variable importance(rfe)
SMI	5.56 ± 1.00	5.62 ± 1.00	5.36 ± 0.99	0.022	12.02	10.63
ASM	13.51 ± 3.73	13.80 ± 3.76	12.58 ± 3.51	0.004	11.01	10.36
BMI	18.61 ± 1.67	18.55 ± 1.67	18.79 ± 1.67	0.198	8.82	7.65
SPPB	7.82 ± 1.66	7.85 ± 1.58	7.72 ± 1.90	0.498	5.22	7.32
Depression	9.78 ± 5.27	9.77 ± 5.15	9.80 ± 5.67	0.956	7.30	6.40
Diabetes	10/430(2.3)	5/329(1.5)	5/101(5.0)	0.104	1.51	7.52
Residence				0.002	3.44	6.96
Rural	260/430(60.5)	212/329()	48/101()
Urban	170/430(39.5)	117/329()	53/101()

Model establishment

A 4-layer feed-forward neural network was constructed using DL, consisting of an input layer, 2 hidden layers, and an output layer, sigmoid as an activation function for neural nodes, constituting our MCI prediction model. 430 patients with sarcopenia were divided into a train set and a test set in a ratio of 8:2 for training and testing. The Loss curve shows that the DL model is well-fitted and does not appear to be under-trained or over-fitted. The loss curve for the train and test set is shown in Fig. 2. Due to the lack of control, we constructed a logistic regression model using the same variables.

Performance of the prediction model

We use the AUC value of the area under the ROC curve to evaluate the model's predictive performance. Our DL prediction model has an AUC of 0.808 (95% CI: 0.704–0.899) on the test set. The model's accuracy was 0.698, precision was 0.956, recall was 0.657, and F1 was 0.778. Logistic regression has an AUC value 0.677 (0.95 CI, 0.548–0.806) on the test set. The ROC curves for the test set are shown in Fig. 2.

Risk stratification

We further calculated the risk for each individual in the entire training cohort. All patients were divided into three groups based on 40% and 60% risk thresholds. A total of 64, 236, and 130 sarcopenia patients were classified into low-risk, medium-risk, and high-risk groups, respectively, with the actual risk probability of critical illness events at 5.38%, 25%, and 54.7%. There was a statistically significant difference between the low-risk, medium-risk, and high-risk groups (P = 0.000). According to the DL model, the false-positive probability of onset in patients with sarcopenia at moderate to high risk of MCI in the next four years was only 1.63%, demonstrating the importance of the DL model in determining whether to intervene.

Validation of the prediction model

Validation was performed using 167 patients with sarcopenia from the CHARLS cohort 2015–2018 who met inclusion criteria, of which 36 had a positive outcome. The AUC of the validation centralized model was 0.755 (0.95 CI, 0.665–0.834). The model's accuracy was 0.683, precision was 0.924, recall was 0.656, and F1 was 0.768. The DL model showed relatively good performance in an independent validation set. The ROC curve for the validation set is shown in Fig. 2.

An online tool

To facilitate clinical application, we have developed an online calculation tool (http://47.115.214.16:5000/). The future MCI risk is calculated based on physical measures and other characteristics of patients with sarcopenia, thus alerting early intervention. The online tool is shown in Fig. 3. The DL model demonstrates the feasibility and utility of MCI prediction in patients with sarcopenia.

All codes for feature selection and model construction etc. are detailed in the supplementary material.

Sarcopenia and MCI are never two parallel lines. They share numerous common pathogenic mechanisms and causative factors. A systematic evaluation of the prevalence of sarcopenia and MCI concluded that the overall prevalence of MCI in patients with sarcopenia was 20.5%, with high heterogeneity. In contrast, the overall prevalence of sarcopenia in patients with MCI was 9.1% [28], which suggests that the prevalence of MCI in patients with sarcopenia is relatively high and that sarcopenia may be a risk factor for MCI. Secondly, several longitudinal studies have shown a significant association between sarcopenia and MCI, and the components of sarcopenia, especially grip strength and gait speed, can be used as predictors of MCI for early prediction and diagnosis of the disease. Therefore, we mainly utilized the body measurements related to sarcopenia for risk prediction of MCI. The variables included were the indicators associated with the onset of MCI, confirmed in previous studies and easily measured daily. Since our model was biased towards clinical consultation, the indicators of complex laboratory tests related to the occurrence of MCI were not included in our inclusion.

Our prediction of MCI in the specific population of sarcopenia, in addition to the high prevalence of MCI, is related to dementia, a severe adverse outcome caused by MCI. First, studies have shown that effective post-intervention can halve the risk of MCI in the next five years in patients with sarcopenia [11], which demonstrates the importance of screening people at high risk for MCI and providing early prevention. Second, 14.4%-55.6% of patients with MCI may regain neurologic integrity [3], suggesting that early diagnosis of MCI is critical. Risk screening and early diagnosis and intervention may significantly reduce the incidence of dementia and reduce the medical and social burden. Cognitive impairment due to sarcopenia differs from other causes or primary cognitive impairment in that they have a particular physical burden that may not be amenable to interventions such as exercise to prevent or mitigate the progression of the disease. Therefore, we hope to use our model to accurately and promptly determine the risk of developing MCI. Further studies can set up personalized intervention programs for patients with sarcopenia who are at high risk of developing MCI and dementia to achieve precise intervention.

Regarding variable inclusion in the model, age and gender are considered variables strongly associated with the onset of MCI. The reason that age was not included in the model may be related to the more concentrated age of the sample. Factors such as ASM and SMI were also included in the model and were calculated from variables such as age and gender. At another level, age, gender, etc., although not included in the DL model, their interactions with other variables are also crucial features that make up the model. Secondly, studies have shown that grip strength and gait speed are also significant predictors of MCI, and the reason for not being included in the model may be related to the definition of sarcopenia. According to the AWGS criteria, muscle mass is used as the primary condition for diagnosing sarcopenia, and muscle strength or physical performance can be diagnosed as sarcopenia if only one is fulfilled. Hence, indicators such as grip strength and gait speed are more variable among individuals, and the model can then not screen out these variables as essential variables. However, based on the definition of severe sarcopenia, where low muscle mass, grip strength, and physical performance co-exist, grip strength and gait speed will probably be essential variables in the model. We were unable to construct a model to predict severe sarcopenia due to the limited nature of our sample, and this will be the focus of our research in the future as well. Third, diabetes itself can lead to cognitive impairment. In patients with sarcopenia, diabetes may be a risk factor for accelerated MCI or dementia.

DL is a new research direction in ML that learns the intrinsic patterns and levels of representation of sample data. Neural networks discover distributed feature representations of data by combining low-level features to form more abstract high-level representations of attribute categories or features. Its essential feature is to try to mimic the pattern of transmitting and processing information between neurons in the brain by designing and establishing appropriate neuron computation nodes and multi-computing hierarchies, selecting proper input and output layers, and through learning and tuning of the network, establishing the input-to output functional relationship. The purpose of our choice of DL modeling is to take advantage of neural networks to learn and infer higher-order nonlinear associations between clinical features and patient outcomes in an entirely data-driven manner [29] and to deeply analyze the effects of variables and interactions between variables on outcomes, which is something that cannot be achieved by ordinary regression, and also, as shown in the results, compared with the basic regression, the feed-forward neural network does show better prediction results. In addition, with the continuous development of DL, its application in the medical industry is becoming more and more extensive. In the future, we can also use DL methods to develop disease-personalized interventions and care for patients.

The reason for the relatively significant difference in AUC values between the base regression model and the DL model may be the complexity of the neural network. The neural network can decrease the error along the gradient by the strength of the connection between the input node and the hidden node, the strength of the connection between the hidden node and the output node, and the threshold value. After repeated training and learning, the weight and threshold value corresponding to the minimum error can be determined. In addition, compared to single regression, neural networks can predict the effect of individual variables on the outcome and the effect of interactions between variables on the MCI.

Our strengths are that the DL model showed good performance in both model construction and validation, and our cohort samples are all over the country, which is also well generalized in China. Second, we used ML methods to double-validate the inclusion variables and performed model construction by combining DL and ML. Third, when searching for domestic and international studies, we could not find other studies that predicted MCI in patients with sarcopenia. Lastly, we built an online computational webpage to facilitate the application of the model. The study also has more shortcomings. First, we lacked foreign validation sets to validate the model and could not prove that the model applies to other races and populations abroad. We are actively searching for data that can be utilized in different countries and regions for validation to solve this problem as early as possible. Second, limited by public databases, including variables related to the onset of MCI, such as the exercise situation, is not comprehensive enough due to the large missing amount and not being included in the study. Third, regarding the assessment of variables, it is generally measured using BIA or DXA, but the formula we used for ASM calculation was also in better agreement with DXA. For walking speed, CHARLS uses measurements at a standard distance of five meters, whereas the international standard is a distance of six meters. To address this issue, it has also been documented that this distance does not affect walking speed [30]. Finally, since the included samples were all over 45 years of age, the DL model is mainly applicable to the elderly population and is not relevant to the prediction of all age groups.

We used CHARLS 2011–2015 data for model construction, ultimately incorporating seven double-validated variables into the model. Compared to logistic regression (AUC: 0.677,0.95 CI: 0.548–0.806), the DL model showed good predictive performance, with an AUC of 0.808 (95% CI: 0.704–0.899) for the test set. We further calculated the risk of morbidity for each individual in the train set and divided all patients into three groups based on 40% and 60% risk probability. There was a statistically significant difference in the risk of morbidity in the low-risk, intermediate-risk, and high-risk groups (P = 0.000). Besides, the model showed a false-positive rate for MCI of 1.63% in intermediate-risk and high-risk sarcopenia populations, suggesting that those populations should have immediate interventions to reduce the unnecessary burden. We used data from CHARLS 2015–2018 for independent validation, and the AUC value for the validation set was 0.76 (95% CI: 0.67–0.83). The DL model has demonstrated the feasibility and utility of predicting the risk of MCI in patients with sarcopenia over the next 4 years (http://47.115.214.16:5000/). It has clinical benefits in identifying sarcopenia patients at different MCI mortality risks.

Author contributions

Xinyue Liu (Formal analysis; Methodology; Writing – original draft; Writing – review & editing); Jingyi Ni (Formal analysis; Writing – review & editing); Baicheng Wang (Methodology; Project administration; Supervision) Rui Yin (Software; Formal analysis); Jinlin Tang (Formal analysis; Supervision); Qi Chu (Methodology; Software; Supervision); Lianghui you (Funding acquisition; Methodology; Project administration; Resources; Supervision); Zhenggang Wu (Formal analysis; Funding acquisition; Investigation); Yan Cao (Methodology; Project administration; Resources); Chenbo Ji (Conceptualization; Formal analysis; Funding acquisition; Methodology; Project administration; Supervision; Writing – review & editing).

Acknowledgments

The authors wish to thank the participants involved in this study and China Social Science Survey Center for their support. Thanks to all the funds that provided support for this article.

Funding

This work was funded by grants from the National Natural Science Foundation of China [82170823]；Nanjing Medical Science and Technique Development Foundation [YKK21166，ZKX23042]；333 High Level Talents Training Project of Jiangsu Province and Science；Jiangsu Association for Science and Technology Youth Science and Technology Talents Lifting Project [TJ-2022-005]; the Key Research and Development Program of Jiangsu Province [BE2021614, JQX22009].

Conflict of interest

The authors have no conflict of interest to report.

Data availability

All the data and material can be available. The data used to support the findings of this study are available from the corresponding author upon request.

Supplementary material

The supplementary material is available in this article.

Jongsiriyanyong S, Limpawattana P (2018) Mild Cognitive Impairment in Clinical Practice: A Review Article. Am J Alzheimers Dis Other Demen 33(8):500-507. https://doi.org/10.1177/1533317518791401
Chen X, Cao M, Liu M et al (2022) Association between sarcopenia and cognitive impairment in the older people: a meta-analysis. Eur Geriatr Med 13(4):771-787. https://doi.org/10.1007/s41999-022-00661-1
Petersen RC, Lopez O, Armstrong MJ et al (2018) Practice guideline update summary: mild cognitive impairment: report of the guideline development, dissemination, and implementation subcommittee of the American academy of neurology. Neurology 90(3):126-135. https://doi.org/10.1212/WNL.0000000000004826
Hu Y, Peng W, Ren R et al (2022) Sarcopenia and mild cognitive impairment among elderly adults: The first longitudinal evidence from CHARLS. J Cachexia Sarcopenia Muscle 13(6):2944-2952. https://doi.org/10.1002/jcsm.13081
Winblad B, Palmer K, Kivipelto M et al (2004) Mild cognitive impairment–beyond controversies, towards a consensus: report of the International Working Group on Mild Cognitive Impairment. J Intern Med 256(3):240-6. https://doi.org/10.1111/j.1365-2796.2004.01380.x
Cruz-Jentoft AJ, Bahat G, Bauer J et al (2019) Sarcopenia: revised European consensus on definition and diagnosis. Age Ageing 48(1):16-31. https://doi.org/10.1093/ageing/afy169
Nishiguchi S, Yamada M, Shirooka H et al (2016) Sarcopenia as a Risk Factor for Cognitive Deterioration in Community-Dwelling Older Adults: A 1-Year Prospective Study. J Am Med Dir Assoc 17(4):372.e5-8. https://doi.org/10.1016/j.jamda.2015.12.096
Ishii H, Makizako H, Doi T et al (2019) Associations of Skeletal Muscle Mass, Lower-Extremity Functioning, and Cognitive Impairment in Community-Dwelling Older People in Japan. J Nutr Health Aging 23(1):35-41. https://doi.org/10.1007/s12603-018-1110-9
Lee D, Kim M, Won CW (2022) Common and different characteristics among combinations of physical frailty and sarcopenia in community-dwelling older adults: The Korean frailty and aging cohort study. Geriatr Gerontol Int 22(1):42-49. https://doi.org/10.1111/ggi.14314
Ramoo K, Hairi NN, Yahya A et al (2022) Longitudinal association between sarcopenia and cognitive impairment among older adults in rural Malaysia. Int J Environ Res Public Health 19(8):4723. https://doi.org/10.3390/ijerph19084723
O'Donovan G, Sarmiento OL, Hessel P et al (2022) Associations of body mass index and sarcopenia with screen-detected mild cognitive impairment in older adults in Colombia. Front Nutr 9:1011967. https://doi.org/10.3389/fnut.2022.1011967
Wu B, Lyu YB, Cao ZJ et al (2021) Associations of sarcopenia, handgrip strength and calf circumference with cognitive impairment among Chinese older adults. Biomed Environ Sci 34(11):859-870. https://doi.org/10.3967/bes2021.119
Kim M, Won CW (2019) Sarcopenia is associated with cognitive impairment mainly due to slow gait speed: results from the Korean Frailty and Aging Cohort Study (KFACS). Int J Environ Res Public Health 16(9):1491. https://doi.org/10.3390/ijerph16091491
Beeri MS, Leugrans SE, Delbono O et al (2021) Sarcopenia is associated with incident Alzheimer's dementia, mild cognitive impairment, and cognitive decline. J Am Geriatr Soc 69(7):1826-1835. https://doi.org/10.1111/jgs.17206
LeCun Y, Bengio Y, Hinton G (2015) Deep learning. J Nature 521(7553):436-44.
Zhao Y, Hu Y, Smith JP et al (2014) Cohort profile: the China Health and Retirement Longitudinal Study (CHARLS). Int J Epidemiol 43(1):61-8. https://doi.org/10.1093/ije/dys203
Chen LK, Woo J, Assantachai P et al (2020) Asian Working Group for Sarcopenia: 2019 Consensus Update on Sarcopenia Diagnosis and Treatment. J Am Med Dir Assoc 21(3):300-307.e2. https://doi.org/10.1016/j.jamda.2019.12.012
Wen X, Wang M, Jiang CM et al (2011) Anthropometric equation for estimation of appendicular skeletal muscle mass in Chinese adults. Asia Pac J Clin Nutr 20(4):551-6.
Yang M, Hu X, Wang H et al (2017) Sarcopenia predicts readmission and mortality in elderly patients in acute care wards: a prospective study. J Cachexia Sarcopenia Muscle 8(2):251-258. https://doi.org/10.1002/jcsm.12163
Delmonico MJ, Harris TB, Lee JS et al (2007) Health, Aging and Body Composition Study. Alternative definitions of sarcopenia, lower extremity performance, and functional impairment with aging in older men and women. J Am Geriatr Soc 55(5):769-74. https://doi.org/10.1111/j.1532-5415.2007.01140.x
Guralnik JM, Simonsick EM, Ferrucci L et al (1994) A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol 49(2):M85-94. https://doi.org/10.1093/geronj/49.2.m85
Crimmins EM, Kim JK, Langa KM et al (2011) Assessment of cognition using surveys and neuropsychological assessment: the Health and Retirement Study and the Aging, Demographics, and Memory Study. J Gerontol B Psychol Sci Soc Sci 66 Suppl 1(Suppl 1):i162-71. https://doi.org/10.1093/geronb/gbr048
Cao L, Zhao Z, Ji C et al (2021) Association between solid fuel use and cognitive impairment: A cross-sectional and follow-up study in a middle-aged and older Chinese population. Environ Int 146:106251. https://doi.org/10.1016/j.envint.2020.106251
Richards M, Touchon J, Ledesert B et al (1999) Cognitive decline in ageing: are AAMI and AACD distinct entities? Int J Geriatr Psychiatry 14(7):534-40. https://doi.org/10.1002/(sici)1099-1166(199907)14:7<534::aid-gps963>3.0.co;2-b
Poon PKM, Tam KW, Zhang D et al (2022) Handgrip strength but not SARC-F score predicts cognitive impairment in older adults with multimorbidity in primary care: a cohort study. BMC Geriatr 22(1):342. https://doi.org/10.1186/s12877-022-03034-2
McGrath R, Vincent BM, Hackney KJ et al (2020) The Longitudinal Associations of Handgrip Strength and Cognitive Function in Aging Americans. J Am Med Dir Assoc 21(5):634-639.e1. https://doi.org/10.1016/j.jamda.2019.08.032
Su D, Chen Z, Chang J et al (2020) Effect of Social Participation on the Physical Functioning and Depression of Empty-Nest Elderly in China: Evidence from the China Health and Retirement Longitudinal Survey (CHARLS). Int J Environ Res Public Health 17(24):9438. https://doi.org/10.3390/ijerph17249438
Yang Y, Xiao M, Leng L et al (2023) A systematic review and meta-analysis of the prevalence and correlation of mild cognitive impairment in sarcopenia. J Cachexia Sarcopenia Muscle 14(1):45-56. https://doi.org/10.1002/jcsm.13143
Liang W, Yao J, Chen A et al (2020) Early triage of critically ill COVID-19 patients using deep learning. Nat Commun 11(1):3543. https://doi.org/10.1038/s41467-020-17280-8
Wu X, Li X, Xu M et al (2021) Sarcopenia prevalence and associated factors among older Chinese population: Findings from the China Health and Retirement Longitudinal Study. PLoS One 16(3):e0247617. https://doi.org/10.1371/journal.pone.0247617

No competing interests reported.

supplementarymaterial.docx

Download PDF

Reviewers agreed at journal
02 Sep, 2024
Reviewers agreed at journal
01 Sep, 2024
Reviewers invited by journal
28 Aug, 2024
Editor assigned by journal
23 Aug, 2024
Submission checks completed at journal
20 Aug, 2024
First submitted to journal
18 Aug, 2024

You are reading this latest preprint version

A relatively accurate prediction model for the risk of developing mild cognitive impairment in patients with sarcopenia: Evidence from the CHARLS

Status:

Version 1

Abstract

Background

Aims

Methods

Results

Conclusions

Figures

Introduction

Methods

Participants

Assessment of sarcopenia

Assessment of MCI

Covariates

Statistical analysis

Results

Demographic characteristics of participants

Selection of predictors

Model establishment

Performance of the prediction model

Risk stratification

Validation of the prediction model

An online tool

Discussion

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1