scholarly journals Development and validation of risk prediction model for cognitive impairment in Chinese community dwellers with normal cognition: using machine learning approach (Preprint)

2020 ◽  
Author(s):  
Mingyue Hu ◽  
Xinyin Wu ◽  
Xinhui Shu ◽  
Yinan Zhao ◽  
Hui Feng
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
Cognitive Impairment ◽  
Cognitive Function ◽  
Older People ◽  
Prediction Model ◽  
Develop Model ◽  
Chinese Elderly ◽  
Normal Cognition ◽  
Community Dwellers

BACKGROUND Dementia causes huge pressure on families and goverments worldwide. Early detection of individuals at risk of cognitive impairment is critical to reduce the mortality rate. OBJECTIVE We aimed to build a prediction model based on machine learning for cognitive impairment (CI) in Chinese elderly community dwellers with normal cognition. METHODS A prospective cohort of 6,718 older people from the Chinese Longitudinal Healthy Longevity Survey (CLHLS), followed from 2008-2011, was used to develop and validate the prediction model. CI was identified using Chinese version of the Mini-Mental State Examination (CMMSE). Several machine learning algorithms (Random Forest, XGBoost, Naïve Bayes and logistic regression) were used to model 3-year risk of CI. We explored optimal cutoffs and adjusted parameters in validation data and evaluated the model in test data. Nomogram was established to vividly present the prediction model. RESULTS Mean age was 80.4 ± 10.3 years and 50.8% were female. During 3-year follow-up, 991 (14.8%) participants were identified as CI. Four features were finally selected to develop model, including age, IADL, marital status and baseline cognitive function. The Concordance index of the model constructed by logistic regression were 0.814 (95%CI 0.781 - 0.846). Older people with normal cognitive function who had a nomogram score of less than 170 or 170 or greater were considered to have low or high 3-year risks of CI, respectively. CONCLUSIONS The simple and feasible CI prediction model could identify Chinese elderly community dwellers at greatest risk for CI. This practical model presented by nomogram could be used to screen Chinese elderly community dwellers for CI and to target intervention strategies.

scholarly journals A Risk Prediction Model Based on Machine Learning for Cognitive Impairment Among Chinese Community-Dwelling Elderly People With Normal Cognition: Development and Validation Study

10.2196/20298 ◽  
2021 ◽  
Vol 23 (2) ◽  
pp. e20298
Author(s):  
Mingyue Hu ◽  
Xinhui Shu ◽  
Gang Yu ◽  
Xinyin Wu ◽  
Maritta Välimäki ◽  
...  
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
Cognitive Impairment ◽  
Older People ◽  
Prediction Model ◽  
Elderly People ◽  
Community Dwelling ◽  
Chinese Community ◽  
Model Based ◽  
Normal Cognition

Background Identifying cognitive impairment early enough could support timely intervention that may hinder or delay the trajectory of cognitive impairment, thus increasing the chances for successful cognitive aging. Objective We aimed to build a prediction model based on machine learning for cognitive impairment among Chinese community-dwelling elderly people with normal cognition. Methods A prospective cohort of 6718 older people from the Chinese Longitudinal Healthy Longevity Survey (CLHLS) register, followed between 2008 and 2011, was used to develop and validate the prediction model. Participants were included if they were aged 60 years or above, were community-dwelling elderly people, and had a cognitive Mini-Mental State Examination (MMSE) score ≥18. They were excluded if they were diagnosed with a severe disease (eg, cancer and dementia) or were living in institutions. Cognitive impairment was identified using the Chinese version of the MMSE. Several machine learning algorithms (random forest, XGBoost, naïve Bayes, and logistic regression) were used to assess the 3-year risk of developing cognitive impairment. Optimal cutoffs and adjusted parameters were explored in validation data, and the model was further evaluated in test data. A nomogram was established to vividly present the prediction model. Results The mean age of the participants was 80.4 years (SD 10.3 years), and 50.85% (3416/6718) were female. During a 3-year follow-up, 991 (14.8%) participants were identified with cognitive impairment. Among 45 features, the following four features were finally selected to develop the model: age, instrumental activities of daily living, marital status, and baseline cognitive function. The concordance index of the model constructed by logistic regression was 0.814 (95% CI 0.781-0.846). Older people with normal cognitive functioning having a nomogram score of less than 170 were considered to have a low 3-year risk of cognitive impairment, and those with a score of 170 or greater were considered to have a high 3-year risk of cognitive impairment. Conclusions This simple and feasible cognitive impairment prediction model could identify community-dwelling elderly people at the greatest 3-year risk for cognitive impairment, which could help community nurses in the early identification of dementia.

scholarly journals Changes in Cognitive Function and Risk Factors for Cognitive Impairment of the Elderly in China: 2005–2014

2019 ◽  
Vol 16 (16) ◽  
pp. 2847 ◽  
Author(s):  
Qilin Zhang ◽  
Yanli Wu ◽  
Tiankuo Han ◽  
Erpeng Liu
Keyword(s):  
Risk Factors ◽  
Logistic Regression ◽  
Regression Analysis ◽  
Cognitive Impairment ◽  
Cognitive Function ◽  
Logistic Regression Analysis ◽  
Binary Logistic Regression ◽  
The Elderly ◽  
Binary Logistic Regression Analysis ◽  
Chinese Elderly

Background: The cognitive function of the elderly has become a focus of public health research. Little is known about the changes of cognitive function and the risk factors for cognitive impairment in the Chinese elderly; thus, the purposes of this study are as follows: (1) to describe changes in cognitive function in the Chinese elderly from 2005–2014 and (2) to explore risk factors for cognitive impairment of the Chinese elderly. Design and setting: A total of 2603 participants aged 64 years and above participated in the Chinese Longitudinal Healthy Longevity Survey (CLHLS) and were followed up from 2005 to 2014. Cognitive function and cognitive impairment were assessed using the Chinese version of the Mini-Mental State Examination (MMSE). Binary logistic regression analysis was used to estimate the odds ratio (OR) and 95% confidence intervals (CI) of cognitive impairment. Results: Results revealed that the cognitive function of the Chinese elderly shows diversified changes: deterioration (55.09%), unchanged (17.21%) and improvement (27.70%). In addition, there are significant demographic differences in gender, age, education, marriage and other aspects when it comes to the changes of cognitive function in Chinese elderly. In the binary logistic regression analysis, female, increased age, lower education level, no spouse, less income, worse PWB (psychological well-being), less fresh fruit and vegetable intake, more activities of daily living (ADL) limitations, lower social engagement were significantly associated with higher odds for cognitive impairment. Conclusions: Various interventions should be implemented to maintain cognitive function in Chinese elderly.

Association of Neutrophil-Lymphocyte Ratio with Mild Cognitive Impairment in Elderly Chinese Adults: A Case-control Study

2020 ◽  
Vol 16 (14) ◽  
pp. 1309-1315
Author(s):  
Peilin An ◽  
Xuan Zhou ◽  
Yue Du ◽  
Jiangang Zhao ◽  
Aili Song ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Cognitive Impairment ◽  
Mild Cognitive Impairment ◽  
Cognitive Function ◽  
Predictive Values ◽  
Neutrophil Lymphocyte Ratio ◽  
Lymphocyte Ratio ◽  
Elderly Chinese ◽  
Normal Cognitive Function ◽  
The Absolute

Background: Inflammation plays a significant role in the pathophysiology of cognitive impairment in previous studies. Neutrophil-lymphocyte ratio (NLR) is a reliable measure of systemic inflammation. Objective: The aim of this study was to investigate the association between NLR and mild cognitive impairment (MCI), and further to explore the diagnostic potential of the inflammatory markers NLR for the diagnosis of MCI in elderly Chinese individuals. Methods: 186 MCI subjects and 153 subjects with normal cognitive function were evaluated consecutively in this study. Neutrophil (NEUT) count and Lymphocyte (LYM) count were measured in fasting blood samples. The NLR was calculated by dividing the absolute NEUT count by the absolute LYM count. Multivariable logistic regression was used to evaluate the potential association between NLR and MCI. NLR for predicting MCI was analyzed using Receiver Operating Characteristic (ROC) curve analysis. Results: The NLR of MCI group was significantly higher than that of subjects with normal cognitive function (2.39 ± 0.55 vs. 1.94 ± 0.51, P < 0.001). Logistic regression analysis showed that higher NLR was an independent risk factor for MCI (OR: 4.549, 95% CI: 2.623-7.889, P < 0.001). ROC analysis suggested that the optimum NLR cut-off point for MCI was 2.07 with 73.66% sensitivity, 69.28% specificity, 74.48% Positive Predictive Values (PPV) and 68.36% negative predictive values (NPV). Subjects with NLR ≥ 2.07 showed higher risk relative to NLR < 2.07 (OR: 5.933, 95% CI: 3.467-10.155, P < 0.001). Conclusion: The elevated NLR is significantly associated with increased risk of MCI. In particular, NLR level higher than the threshold of 2.07 was significantly associated with the probability of MCI.

scholarly journals Effect of nutritional risk on cognitive function in patients with chronic obstructive pulmonary disease

2021 ◽  
Vol 49 (2) ◽  
pp. 030006052199012
Author(s):  
Yiben Huang ◽  
Jiedong Ma ◽  
Bingqian Jiang ◽  
Naiping Yang ◽  
Fangyi Fu ◽  
...  
Keyword(s):  
Chronic Obstructive Pulmonary Disease ◽  
Logistic Regression ◽  
Regression Analysis ◽  
Cognitive Impairment ◽  
Cognitive Function ◽  
Pulmonary Disease ◽  
Nutritional Risk ◽  
Chronic Obstructive ◽  
Obstructive Pulmonary Disease ◽  
Nrs 2002

Objective We aimed to clarify the cognitive function of patients with chronic obstructive pulmonary disease (COPD) and different nutritional status. Methods Among 95 patients with COPD in this retrospective study, we administered the Nutritional Risk Screening 2002 (NRS 2002) and Mini-Mental State Examination (MMSE). We recorded patients’ clinical characteristics, comorbidities, and laboratory measurements. According to NRS 2002 scores, patients were divided into two groups: no nutritional risk with NRS 2002 < 3 ( n = 54) and nutritional risk, with NRS 2002 ≥ 3 ( n = 41). Results We found a negative correlation between NRS 2002 and MMSE scores in participants with COPD ( r = −0.313). Patients with nutritional risk were more likely to be cognitively impaired than those with no nutritional risk. Multivariate logistic regression analysis indicated that malnutrition was an independent risk factor for cognitive impairment, after adjusting for confounders (odds ratio [OR] = 4.120, 95% confidence interval [CI]: 1.072–15.837). We found a similar association between NRS 2002 and MMSE scores at 90-day follow-up using a Pearson’s correlation test ( r = −0.493) and logistic regression analysis (OR = 7.333, 95% CI: 1.114–48.264). Conclusions Patients with COPD at nutritional risk are more likely to have cognitive impairment.

Cross-Sectional Examination of Homocysteine Levels with Sarcopenia and Its Components in Memory Clinic Outpatients

2021 ◽  
pp. 1-10
Author(s):  
Yosuke Yamada ◽  
Hiroyuki Umegaki ◽  
Fumie Kinoshita ◽  
Chi Hsien Huang ◽  
Taiki Sugimoto ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Regression Analysis ◽  
Cognitive Impairment ◽  
Cognitive Function ◽  
Linear Regression ◽  
Logistic Regression Analysis ◽  
Linear Regression Analysis ◽  
Multivariate Logistic Regression Analysis ◽  
Memory Clinic ◽  
Cross Sectional

Background: Homocysteine is a common risk factor for cognitive impairment and sarcopenia. However, very few studies have shown an association between sarcopenia and serum homocysteine levels after adjustment for cognitive function. Objective: The purpose of this study was to investigate the relationship between homocysteine and sarcopenia in memory clinic patients. Methods: This cross-sectional study investigated outpatients in a memory clinic. We enrolled 1,774 participants (≥65 years old) with measured skeletal muscle mass index (SMI), hand grip strength (HGS), and homocysteine. All participants had undergone cognitive assessments and were diagnosed with dementia, mild cognitive impairment, or normal cognition. Patient characteristics were compared according to sarcopenia presence, SMI level, or HGS. Multivariate logistic regression analysis was performed to determine the association of homocysteine with sarcopenia, low SMI, or low HGS. Next, linear regression analysis was performed using HGS as a continuous variable. Results: Logistic regression analysis showed that low HGS was significantly associated with homocysteine levels (p = 0.002), but sarcopenia and low SMI were not. In linear regression analysis, HGS was negatively associated with homocysteine levels after adjustment for Mini-Mental State Examination score (β= –2.790, p <  0.001) or clinical diagnosis of dementia (β= –3.145, p <  0.001). These results were similar for men and women. Conclusion: Our results showed a negative association between homocysteine and HGS after adjustment for cognitive function. Our findings strengthen the assumed association between homocysteine and HGS. Further research is needed to determine whether lower homocysteine levels lead to prevent muscle weakness.

Abstract MP003: Development and Validation of a Prediction Model of Weight Loss Maintenance or Regain at 4 Years in the Look AHEAD Trial

Circulation ◽  
2017 ◽  
Vol 135 (suppl_1) ◽  
Author(s):  
Samantha E Berger ◽  
Gordon S Huggins ◽  
Jeanne M McCaffery ◽  
Alice H Lichtenstein
Keyword(s):  
Machine Learning ◽  
Type 2 Diabetes ◽  
Weight Loss ◽  
Logistic Regression ◽  
Prediction Model ◽  
Weight Regain ◽  
Initial Weight ◽  
Initial Weight Loss ◽  
Look Ahead

Introduction: The development of type 2 diabetes is strongly associated with excess weight gain and can often be partially ameliorated or reversed by weight loss. While many lifestyle interventions have resulted in successful weight loss, strategies to maintain the weight loss have been considerably less successful. Prior studies have identified multiple predictors of weight regain, but none have synthesized them into one analytic stream. Methods: We developed a prediction model of 4-year weight regain after a one-year lifestyle-induced weight loss intervention followed by a 3 year maintenance intervention in 1791 overweight or obese adults with type 2 diabetes from the Action for Health in Diabetes (Look AHEAD) trial who lost ≥3% of initial weight by the end of year 1. Weight regain was defined as regaining <50% of the weight lost during the intervention by year 4. Using machine learning we integrated factors from several domains, including demographics, psychosocial metrics, health status and behaviors (e.g. physical activity, self-monitoring, medication use and intervention adherence). We used classification trees and stochastic gradient boosting with 10-fold cross validation to develop and internally validate the prediction model. Results: At the end of four years, 928 individuals maintained ≥50% of their initial weight lost (maintainers), whereas 863 did not met that criterion (regainers). We identified an interaction between age and several variables in the model, as well as percent initial weight loss. Several factors were significant predictors of weight regain based on variable importance plots, regardless of age or initial weight loss, such as insurance status, physical function score, baseline BMI, meal replacement use and minutes of exercise recorded during year 1. We also identified several factors that were significant predictors depending on age group (45-55y/ 56-65y/66-76y) and initial weight loss (lost 3-9% vs. ≥10% of initial weight). When the variables identified from machine learning were added to a logistic regression model stratified by age and initial weight loss groups, the models showed good prediction (3-9% initial weight loss, ages 45-55y (n=293): ROC AUC=0.78; ≥10% initial weight loss, ages 45-55y (n=242): ROC AUC=0.78; (3-9% initial weight loss, ages 56-65y (n=484): ROC AUC=0.70; ≥10% initial weight loss, ages 56-65y (n=455): ROC AUC = 0.74; 3-9% initial weight loss, ages 66-76y (n=150): ROC AUC=0.84; ≥10% initial weight loss, ages 66-76y (n=167): ROC AUC=0.86). Conclusion: The combination of machine learning methodology and logistic regression generates a prediction model that can consider numerous factors simultaneously, can be used to predict weight regain in other populations and can assist in the development of better strategies to prevent post-loss regain.

scholarly journals Increasing prevalence of anticholinergic medication use in older people in England over 20 years: Cognitive Function and Ageing Study I and II

2019 ◽  
Author(s):  
Carlota Grossi ◽  
Kathryn Richardson ◽  
George Savva ◽  
Chris Fox ◽  
Antony Arthur ◽  
...  
Keyword(s):  
Cognitive Impairment ◽  
Cognitive Function ◽  
Older People ◽  
Mini Mental State Examination ◽  
Medication Use ◽  
Anticholinergic Medication ◽  
Older Population ◽  
Clinically Significant ◽  
State Examination ◽  
Anticholinergic Burden

Abstract Background: Anticholinergic medication use is linked with increased cognitive decline, dementia, falls and mortality. The characteristics of the population who use anticholinergic medication are not known. Here we estimate the prevalence of anticholinergic use in England’s older population in 1991 and 2011, and describe changes in use by participant’s age, sex, cognition and disability. Methods: We compared data from participants aged 65+ years from the Cognitive Function and Ageing Studies (CFAS I and II), collected during 1990-1993 (N=7,635) and 2008-2011 (N=7,762). We estimated the prevalence of potent anticholinergic use (Anticholinergic Cognitive Burden [ACB] score=3) and average anticholinergic burden (sum of ACB scores), using inverse probability weights standardised to the 2011 UK population. These were stratified by age, sex, Mini-Mental State Examination score, and activities of daily living (ADL) or instrumental ADL (IADL) disability. Results: Prevalence of potent anticholinergic use increased from 5.7% (95% Confidence Interval [CI] 5.2-6.3%) of the older population in 1990-93 to 9.9% (9.3-10.7%) in 2008-11, adjusted odds ratio of 1.90 (95%CI 1.67 – 2.16). People with clinically significant cognitive impairment (MMSE [Mini Mental State Examination] 21 or less) were the heaviest users of potent anticholinergic in CFAS II (16.5% [95%CI 12.0-22.3%]). Large increases in the prevalence of the use medication with ‘any’ anticholinergic activity were seen in older people with clinically significant cognitive impairment (53.3% in CFAS I to 71.5% in CFAS II). Conclusions: Use of potent anticholinergic medications nearly doubled in England’s older population over 20 years with some of the greatest increases amongst those particularly vulnerable to anticholinergic side-effects. Key words: cognitive impairment, anticholinergic burden, polypharmacy

scholarly journals Fire Risk Assessment Models Using Statistical Machine Learning and Optimized Risk Indexing

2020 ◽  
Vol 10 (12) ◽  
pp. 4199
Author(s):  
Myoung-Young Choi ◽  
Sunghae Jun
Keyword(s):  
Machine Learning ◽  
Risk Assessment ◽  
Logistic Regression ◽  
Prediction Model ◽  
Fire Risk ◽  
Fire Occurrence ◽  
Statistical Machine Learning ◽  
Fire Insurance ◽  
Comparison Results ◽  
Fire Risk Assessment

It is very difficult for us to accurately predict occurrence of a fire. But, this is very important to protect human life and property. So, we study fire hazard prediction and evaluation methods to cope with fire risks. In this paper, we propose three models based on statistical machine learning and optimized risk indexing for fire risk assessment. We build logistic regression, deep neural networks (DNN) and fire risk indexing models, and verify performances between proposed and traditional models using real investigated data related to fire occurrence in Korea. In general, fire prediction models currently in use do not provide satisfactory levels of accuracy. The reason for this result is that the factors affecting fire occurrence are very diverse and frequency of fire occurrence is very sparse. To improve accuracy of fire occurrence, we first build logistic regression and DNN models. In addition, we construct a fire risk indexing model for a more improved model of fire prediction. To illustrate comparison results between our research models and current fire prediction model, we use real fire data investigated in Korea between 2011 to 2017. From the experimental results of this paper, we can confirm that accuracy of prediction by the proposed method is superior to the existing fire occurrence prediction model. Therefore, we expect the proposed model to contribute to evaluating the possibility of fire risk in buildings and factories in the field of fire insurance and to calculate the fire insurance premium.

scholarly journals 352. Characteristics Associated with Pre-Frailty in Older People Living with HIV

2019 ◽  
Vol 6 (Supplement_2) ◽  
pp. S186-S186
Author(s):  
Peter Mazonson ◽  
Theoren Loo ◽  
Jeff Berko ◽  
Sarah-Marie Chan ◽  
Ryan Westergaard ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Regression Analysis ◽  
Cognitive Function ◽  
Older People ◽  
Logistic Regression Analysis ◽  
The United States ◽  
People Living With Hiv ◽  
Factors Associated ◽  
Increased Risk ◽  
Living With Hiv

Abstract Background Frailty is a concern among older people living with HIV (PLHIV). There is a paucity of research characterizing PLHIV who are at risk of becoming frail (pre-frailty). To investigate how HIV impacts older PLHIV in the United States, a new study called Aging with Dignity, Health, Optimism and Community (ADHOC) was launched at ten sites to collect self-reported data. This analysis uses data from ADHOC to identify factors associated with pre-frailty. Methods Pre-frailty was assessed using the Frailty Index for Elders (FIFE), where a score of zero indicated no frailty, 1–3 indicated pre-frailty, and 4–10 indicated frailty. A cross-sectional analysis was performed on 262 PLHIV (age 50+) to determine the association between pre-frailty and self-reported sociodemographic, health, and clinical indicators using bivariate analyses. Factors associated with pre-frailty were then included in a logistic regression analysis using backward selection. Results The average age of ADHOC participants was 59 years. Eighty-two percent were male, 66% were gay or lesbian, and 56% were white. Forty-seven percent were classified with pre-frailty, 26% with frailty, and 27% with no frailty. In bivariate analyses, pre-frailty was associated with depression, low cognitive function, depression, multiple comorbidities, low income, low social support and unemployment (Table 1). In the multiple logistic regression analysis, pre-frailty was associated with having low cognitive function (Odds Ratio [OR] 8.56, 95% Confidence Interval [CI]: 3.24–22.63), 4 or more comorbid conditions (OR 4.00, 95% CI: 2.23–7.06), and an income less than $50,000 (OR 2.70, 95% CI: 1.56–4.68) (Table 2). Conclusion This study shows that commonly collected clinical and sociodemographic metrics can help identify PLWH who are more likely to have pre-frailty. Early recognition of factors associated with pre-frailty among PLHIV may help to prevent progression to frailty. Understanding markers of increased risk for pre-frailty may help clinicians and health systems better target multi-modal interventions to prevent negative health outcomes associated with frailty. Disclosures All authors: No reported disclosures.

Sign in / Sign up

Export Citation Format

Share Document