scholarly journals Multidimensional Sleep and Mortality in Older Adults: A Machine-Learning Comparison With Other Risk Factors

2019 ◽  
Vol 74 (12) ◽  
pp. 1903-1909 ◽  
Author(s):  
Meredith L Wallace ◽  
Daniel J Buysse ◽  
Susan Redline ◽  
Katie L Stone ◽  
Kristine Ensrud ◽  
...  
Keyword(s):  
Machine Learning ◽  
Risk Factors ◽  
Older Adults ◽  
Cardiovascular Mortality ◽  
Predictive Ability ◽  
Future Research ◽  
Applied Machine Learning ◽  
Predictive Variables ◽  
Time In Bed

Abstract Background Sleep characteristics related to duration, timing, continuity, and sleepiness are associated with mortality in older adults, but rarely considered in health recommendations. We applied machine learning to: (i) establish the predictive ability of a multidimensional self-reported sleep domain for all-cause and cardiovascular mortality in older adults relative to other established risk factors and (ii) to identify which sleep characteristics are most predictive. Methods The analytic sample includes N = 8,668 older adults (54% female) aged 65–99 years with self-reported sleep characterization and longitudinal follow-up (≤15.5 years), aggregated from three epidemiological cohorts. We used variable importance (VIMP) metrics from a random survival forest to rank the predictive abilities of 47 measures and domains to which they belong. VIMPs > 0 indicate predictive variables/domains. Results Multidimensional sleep was a significant predictor of all-cause (VIMP [99.9% confidence interval {CI}] = 0.94 [0.60, 1.29]) and cardiovascular (1.98 [1.31, 2.64]) mortality. For all-cause mortality, it ranked below that of the sociodemographic (3.94 [3.02, 4.87]), physical health (3.79 [3.01, 4.57]), and medication (1.33 [0.94, 1.73]) domains but above that of the health behaviors domain (0.22 [0.06, 0.38]). The domains were ranked similarly for cardiovascular mortality. The most predictive individual sleep characteristics across outcomes were time in bed, hours spent napping, and wake-up time. Conclusion Multidimensional sleep is an important predictor of mortality that should be considered among other more routinely used predictors. Future research should develop tools for measuring multidimensional sleep—especially those incorporating time in bed, napping, and timing—and test mechanistic pathways through which these characteristics relate to mortality.

scholarly journals Prevalence and Risk Factors of Postprandial Hypotension Among Japanese Older Adults in a Facility

2020 ◽  
Vol 4 (Supplement_1) ◽  
pp. 222-222
Author(s):  
Noriko Suzuki ◽  
Masahiko Hashizume ◽  
Hideyuki Shiotani
Keyword(s):  
Risk Factors ◽  
Blood Pressure ◽  
Older Adults ◽  
Older People ◽  
Future Research ◽  
Baseline Body Mass Index ◽  
Term Care ◽  
Postprandial Hypotension ◽  
Before And After ◽  
Japanese Older Adults

Abstract Postprandial hypotension (PPH) is an unrecognized sudden drop of blood pressure (BP) after meals and a hidden problem among older people including those living in long-term care facilities (LTCFs). Though PPH causes dizziness, falls, and syncope, it has received little attention from¬¬¬ healthcare workers (HCW) including caregivers, nurses and physicians, and risk factors of PPH should be carefully assessed to improve quality of life. Therefore, we aimed to examine the prevalence and risk factors of PPH in a LTCF in Japan. Participants were 114 older adults living in a LTCF in Japan (mean age 85.9 years old; 85 female (74%)). To examine PPH, blood pressure (BP) was measured before and after lunch. BP after meal was measured four times every 30 minutes. PPH is defined as a BP drop of 20 mmHg or more and we also defined a BP drop within a range of 19 to 15 mmHg as potential-PPH. As risk factors, we compared systolic and diastolic BP at baseline, body mass index, pulse rate, disease and complications between groups with/without PPH. The prevalence of PPH was 41% (47/114) and 52% with potential-PPH; 11% (13/114) added. Among risk factors, systolic BP was significantly higher in those with PPH (142.6 vs 123.5 mmHg, p <0.001). This study revealed that PPH & potential-PPH occurred in half of the subjects in a LTCF in Japan. HCW need to focus on high systolic BP to predict PPH and future research is necessary to prevent and cope with PPH for older people.

Animal protein intake is inversely associated with mortality in older adults: the InCHIANTI study

2021 ◽  
Author(s):  
Tomás Meroño ◽  
Raúl Zamora-Ros ◽  
Nicole Hidalgo-Liberona ◽  
Montserrat Rabassa ◽  
Stefania Bandinelli ◽  
...  
Keyword(s):  
Older Adults ◽  
Total Energy ◽  
Cardiovascular Mortality ◽  
Protein Intake ◽  
Cox Regression ◽  
Clinical Information ◽  
Community Dwelling ◽  
Animal Protein ◽  
Plant Protein

Abstract Background In general, plant protein intake was inversely associated with mortality in studies in middle-aged adults. Our aim was to evaluate the long-term associations of animal and plant protein intake with mortality in older adults. Methods A prospective cohort study including 1,139 community-dwelling older adults (mean age 75 years, 56% women) living in Tuscany, Italy, followed for 20 years (InCHIANTI study) was analyzed. Dietary intake by food frequency questionnaires and clinical information were assessed five times during the follow-up. Protein intakes were expressed as percentages of total energy. Time-dependent Cox regression models adjusted for confounders were used to assess the association between plant and animal protein intake, and mortality. Results During the 20-years of follow up (mean: 12y), 811 deaths occurred (292 of cardiovascular- and 151 of cancer-related causes). Animal protein intake was inversely associated with all-cause (HR per 1% of total energy from protein increase, 95%CI: 0.96, 0.93-0.99) and cardiovascular mortality (HR per 1% of total energy from protein increase, 95%CI: 0.93, 0.87-0.98). Plant protein intake showed no association with any of the mortality outcomes, but an interaction with baseline hypertension was found for all-cause and cardiovascular mortality (p<0.05). Conclusions Animal protein was inversely associated with all-cause and cardiovascular mortality in older adults. Further studies are needed to provide recommendations on dietary protein intake for older adults.

Abstract P122: Racial/Ethnic Variation in Incidence and Progression of Coronary Artery Calcification: Implications for Understanding the Hispanic Paradox

Circulation ◽  
2018 ◽  
Vol 137 (suppl_1) ◽  
Author(s):  
Sandra S Albrecht ◽  
Pamela L Lutsey ◽  
Matthew Allison ◽  
Teresa Seeman ◽  
Martha L Daviglus ◽  
...  
Keyword(s):  
Risk Factors ◽  
Coronary Artery ◽  
Future Research ◽  
Foreign Born ◽  
Hispanic Health ◽  
Current Cigarette Smoking ◽  
Current Cigarette ◽  
Traditional Risk Factors ◽  
Adjusted Incidence

Background: Previous studies show that Hispanic persons have similar or lower levels of coronary artery calcium (CAC) and slower progression than non-Hispanic whites (NHW), even after adjustment for traditional risk factors. We examined whether this health advantage in CAC incidence and progression among Hispanic adults extends across all levels of risk factor (RF) burden, and whether associations vary by nativity (foreign-born, US-born) and by heritage group (Mexican, non-Mexican). Methods: We analyzed data on Hispanic and NHW participants aged 45-84 years from the Multi-Ethnic Study of Atherosclerosis (MESA). Follow-up CAC measurements and complete covariate data were available for 3694 participants with an average of 6.6 years between the follow-up and baseline scans (2000-2002). Baseline measures of the following traditional RFs were considered: current cigarette smoking, high total cholesterol, hypertension, diabetes, and obesity, with RF burden scores ranging from 0-5. Outcomes were incident CAC (any follow-up CAC >0 Agatston units) among individuals without detectable CAC at baseline, and CAC progression (any positive increase in CAC) among all participants estimated using relative risk regression. All models were adjusted for age, sex, RF burden, race/ethnicity, education, income, and time between scans Results: Although a higher proportion of Hispanics had RF burden scores ≥3 compared to NHW (14.6% vs 8.9%, p<0.0001), Hispanics had a lower adjusted incidence (risk ratio (RR) = 0.83, 95% CI: 0.72-0.96) and less progression of CAC (RR=0.90, 95% CI: 0.86-0.95) than NHW. However, there was evidence of heterogeneity in this pattern. For example, among individuals with no detectable baseline CAC, a Hispanic health advantage was only seen among individuals with RF burden scores of 0 (RR=0.66, 95% CI: 0.48-0.91 for Hispanics vs. NHW at RF=0), with race/ethnic differences getting progressively smaller with increasing RF burden (for RF ≥3: RR=1.01, 95% CI: 0.69-1.48). Compared to NHW, lower adjusted incidence and progression of CAC was evident to an even greater extent among foreign-born Hispanics, but a health advantage was still present for US-born Hispanics, and for both Hispanic heritage groups. However, these patterns also only remained among individuals with lower RF burden scores. Conclusions: The Hispanic health advantage in CAC incidence and progression was primarily evident among individuals with fewer traditional risk factors for CVD, but was present among different Hispanics groups. Future research is necessary to identify the factors underlying this advantage, and the dynamics that erode it as RF burden increases.

scholarly journals ADVANCING AN INTERDISCIPLINARY SCIENCE OF AGING THROUGH A PRACTICE-BASED DATA SCIENCE APPROACH

2019 ◽  
Vol 3 (Supplement_1) ◽  
pp. S480-S480
Author(s):  
Robert Lucero ◽  
Ragnhildur Bjarnadottir
Keyword(s):  
Machine Learning ◽  
Risk Factors ◽  
Older Adults ◽  
Text Mining ◽  
Language Processing ◽  
Fall Risk ◽  
Data Science ◽  
Care Quality ◽  
Science Approach ◽  
Hospitalized Older Adults

Abstract Two hundred and fifty thousand older adults die annually in United States hospitals because of iatrogenic conditions (ICs). Clinicians, aging experts, patient advocates and federal policy makers agree that there is a need to enhance the safety of hospitalized older adults through improved identification and prevention of ICs. To this end, we are building a research program with the goal of enhancing the safety of hospitalized older adults by reducing ICs through an effective learning health system. Leveraging unique electronic data and healthcare system and human resources at the University of Florida, we are applying a state-of-the-art practice-based data science approach to identify risk factors of ICs (e.g., falls) from structured (i.e., nursing, clinical, administrative) and unstructured or text (i.e., registered nurse’s progress notes) data. Our interdisciplinary academic-clinical partnership includes scientific and clinical experts in patient safety, care quality, health outcomes, nursing and health informatics, natural language processing, data science, aging, standardized terminology, clinical decision support, statistics, machine learning, and hospital operations. Results to date have uncovered previously unknown fall risk factors within nursing (i.e., physical therapy initiation), clinical (i.e., number of fall risk increasing drugs, hemoglobin level), and administrative (i.e., Charlson Comorbidity Index, nurse skill mix, and registered nurse staffing ratio) structured data as well as patient cognitive, environmental, workflow, and communication factors in text data. The application of data science methods (i.e., machine learning and text-mining) and findings from this research will be used to develop text-mining pipelines to support sustained data-driven interdisciplinary aging studies to reduce ICs.

scholarly journals Risk factors for hospital readmission in older adults within 30 days of discharge – a comparative retrospective study

2020 ◽  
Vol 20 (1) ◽  
Author(s):  
Maria Glans ◽  
Annika Kragh Ekstam ◽  
Ulf Jakobsson ◽  
Åsa Bondesson ◽  
Patrik Midlöv
Keyword(s):  
Risk Factors ◽  
Older Adults ◽  
Retrospective Study ◽  
Home Care ◽  
Hospital Readmission ◽  
Patient Characteristics ◽  
Transition Of Care ◽  
Discharge Transition ◽  
Surgical Unit

Abstract Background The area of hospital readmission in older adults within 30 days of discharge is extensively researched but few studies look at the whole process. In this study we investigated risk factors related, not only to patient characteristics prior to and events during initial hospitalisation, but also to the processes of discharge, transition of care and follow-up. We aimed to identify patients at most risk of being readmitted as well as processes in greatest need of improvement, the goal being to find tools to help reduce early readmissions in this population. Methods This comparative retrospective study included 720 patients in total. Medical records were reviewed and variables concerning patient characteristics prior to and events during initial hospital stay, as well as those related to the processes of discharge, transition of care and follow-up, were collected in a standardised manner. Either a Student’s t-test, χ2-test or Fishers’ exact test was used for comparisons between groups. A multiple logistic regression analysis was conducted to identify variables associated with readmission. Results The final model showed increased odds of readmission in patients with a higher Charlson Co-morbidity Index (OR 1.12, p-value 0.002), excessive polypharmacy (OR 1.66, p-value 0.007) and living in the community with home care (OR 1.61, p-value 0.025). The odds of being readmitted within 30 days increased if the length of stay was 5 days or longer (OR 1.72, p-value 0.005) as well as if being discharged on a Friday (OR 1.88, p-value 0.003) or from a surgical unit (OR 2.09, p-value 0.001). Conclusion Patients of poor health, using 10 medications or more regularly and living in the community with home care, are at greater risk of being readmitted to hospital within 30 days of discharge. Readmissions occur more often after being discharged on a Friday or from a surgical unit. Our findings indicate patients at most risk of being readmitted as well as discharging routines in most need of improvement thus laying the ground for further studies as well as targeted actions to take in order to reduce hospital readmissions within 30 days in this population.

scholarly journals Nonimmersive Brain Gaming for Older Adults With Cognitive Impairment: A Scoping Review

2019 ◽  
Vol 59 (6) ◽  
pp. e764-e781 ◽  
Author(s):  
Pallavi Sood ◽  
Sandra L Kletzel ◽  
Shilpa Krishnan ◽  
Hannes Devos ◽  
Ahmed Negm ◽  
...  
Keyword(s):  
Older Adults ◽  
Cognitive Impairment ◽  
Scoping Review ◽  
Research Field ◽  
Cognitive Outcome ◽  
Future Research ◽  
Eligibility Criteria ◽  
Technological Advances ◽  
Design And Methods

Abstract Background Technological advances have allowed a variety of computerized cognitive training tools to be engineered in ways that are fun and entertaining yet challenging at a level that can maintain motivation and engagement. This revolution has created an opportunity for gerontological scientists to evaluate brain gaming approaches to improve cognitive and everyday function. The purpose of this scoping review is to provide a critical overview of the existing literature on nonimmersive, electronic brain gaming interventions in older adults with mild cognitive impairment or dementia. Research Design and Methods Systematic search was conducted using 7 electronic databases from inception through July 2017. A comprehensive 2-level eligibility process was used to identify studies for inclusion based on PRISMA guidelines. Results Seventeen studies met eligibility criteria. Majority of the studies were randomized controlled trials (n = 13) and incorporated an active control (n = 9). Intervention doses ranged from 4 to 24 weeks in duration with an average of 8.4 (±5.1 standard deviation [SD]) weeks. Session durations ranged from 30 to 100 min with an average of 54 (±25 SD) minutes. Nearly half of studies included a follow-up, ranging from 3 months to 5 years (n = 8). For most studies, brain gaming improved at least one cognitive outcome (n = 12); only one study reported improvement in activities of daily living. Discussion and Implications This scoping review conveys the breadth of an emerging research field, which will help guide future research to develop standards and recommendations for brain gaming interventions which are currently lacking.

P1755Plasma proteomics in healthy individuals predict myocardial infarction

2019 ◽  
Vol 40 (Supplement_1) ◽  
Author(s):  
R Hoogeveen ◽  
J P Belo Pereira ◽  
V Zampoleri ◽  
M J Bom ◽  
W Koenig ◽  
...  
Keyword(s):  
Machine Learning ◽  
Risk Factors ◽  
Myocardial Infarction ◽  
Acute Myocardial Infarction ◽  
Plasma Proteins ◽  
Risk Model ◽  
Machine Learning Algorithms ◽  
Machine Learning Techniques ◽  
Traditional Risk Factors

Abstract Background Currently used models to predict cardiovascular event risk have limited value. It has been shown repetitively that the addition of single biomarkers has modest impact. Recently we observed that a model consisting of a larger array of plasma proteins performed very well in predicting the presence of vulnerable plaques in primary prevention patients. However, the validation of this protein panel in predicting cardiovascular outcomes remains to be established. Purpose This study investigated the ability of a 384 preselected protein biomarkers to predict acute myocardial infarction, using state-of-the-art machine learning techniques. Secondly, we compared the performance of this multi-protein risk model to traditional risk engines. Methods We selected 822 subjects from the EPIC-Norfolk prospective cohort study, of whom 411 suffered a myocardial infarction during follow-up (median 15 years) compared to 411 controls who remained event-free (median follow-up 20 years). The 384 proteins were measured using proximity extension assay technology. Machine learning algorithms (random forests) were used for the prediction of acute myocardial infarction (ICD code I21–22). Performance of the model was tested against and on top of traditional risk factors for cardiovascular disease (refit Framingham). All performance measurements were averaged over several stability selection routines. Results Prediction of myocardial infarction using a machine-learning model consisting of 50 plasma proteins resulted in a ROC AUC of 0.74±0.14, in comparison to 0.69±0.17 using traditional risk factors (refit Framingham. Combining the proteins and refit Framingham resulted in a ROC AUC of 0.74±0.15. Focussing on events occurring within 3 years after baseline blood withdrawal, the ROC AUC increased to 0.80±0.09 using 50 plasma proteins, as opposed to 0.67±0.22 using refit Framingham (figure). Combining the protein model with refit Framingham resulted in a ROC AUC of 0.82±0.11 for these events. Diagnostic performance events <3yrs Conclusion High-throughput proteomics outperforms traditional risk factors in prediction of acute myocardial infarction. Prediction of myocardial infarction occurring within 3 years after inclusion showed highest performance. Availability of affordable proteomic approaches and developed machine learning pave the path for clinical implementation of these models in cardiovascular risk prediction. Acknowledgement/Funding This study was funded by an ERA-CVD grant (JTC2017) and EU Horizon 2020 grant (REPROGRAM, 667837)

scholarly journals The Application of Machine Learning to a General Risk–Need Assessment Instrument in the Prediction of Criminal Recidivism

2020 ◽  
pp. 009385482096975
Author(s):  
Mehdi Ghasemi ◽  
Daniel Anvari ◽  
Mahshid Atapour ◽  
J. Stephen wormith ◽  
Keira C. Stockdale ◽  
...  
Keyword(s):  
Machine Learning ◽  
Predictive Accuracy ◽  
Characteristic Curve ◽  
Assessment Instrument ◽  
Support Vector ◽  
Data Set ◽  
Applied Machine Learning ◽  
Vector Machines ◽  
Individual Scores

The Level of Service/Case Management Inventory (LS/CMI) is one of the most frequently used tools to assess criminogenic risk–need in justice-involved individuals. Meta-analytic research demonstrates strong predictive accuracy for various recidivism outcomes. In this exploratory study, we applied machine learning (ML) algorithms (decision trees, random forests, and support vector machines) to a data set with nearly 100,000 LS/CMI administrations to provincial corrections clientele in Ontario, Canada, and approximately 3 years follow-up. The overall accuracies and areas under the receiver operating characteristic curve (AUCs) were comparable, although ML outperformed LS/CMI in terms of predictive accuracy for the middle scores where it is hardest to predict the recidivism outcome. Moreover, ML improved the AUCs for individual scores to near 0.60, from 0.50 for the LS/CMI, indicating that ML also improves the ability to rank individuals according to their probability of recidivating. Potential considerations, applications, and future directions are discussed.

Factors Associated With Multimorbidity Patterns in Older Adults in England: Findings From the English Longitudinal Study of Aging (ELSA)

2019 ◽  
Vol 32 (9) ◽  
pp. 1120-1132 ◽  
Author(s):  
Hai Nguyen ◽  
Kia-Chong Chua ◽  
Alexandru Dregan ◽  
Silia Vitoratou ◽  
Ivet Bayes-Marin ◽  
...  
Keyword(s):  
Risk Factors ◽  
Older Adults ◽  
Longitudinal Study ◽  
Latent Class ◽  
Future Research ◽  
Factors Associated ◽  
Lifestyle Risk Factors ◽  
Class B ◽  
Lifestyle Risk ◽  
English Longitudinal Study

Objective: We aimed to identify the patterns of multimorbidity in older adults and explored their association with sociodemographic and lifestyle risk factors. Method: The sample included 9,171 people aged 50+ from Wave 2 of the English Longitudinal Study of Aging (ELSA). Latent Class Analysis (LCA) was performed on 26 chronic diseases to determine clusters of common diseases within individuals and their association with sociodemographic and lifestyle risk factors. Result: Three latent classes were identified: (a) a cardiorespiratory/arthritis/cataracts class, (b) a metabolic class, and (c) a relatively healthy class. People aged 70 to 79 were 9.91 times (95% Confidence Interval [CI] = [5.13, 19.13]) more likely to be assigned to the cardiorespiratory/arthritis/cataracts class, while regular drinkers and physically inactive people were 0.33 times (95% CI = [0.24, 0.47]) less likely to be assigned to this class. Conclusion: Future research should investigate these patterns further to gain more insights into the needs of people with multimorbidity.

Relationship Between Insomnia, Depression, and Mortality: A 12-Year Follow-Up of Older Adults in the Community

2000 ◽  
Vol 12 (3) ◽  
pp. 295-306 ◽  
Author(s):  
Lena Mallon ◽  
Jan-Erik Broman ◽  
Jerker Hetta
Keyword(s):  
Risk Factors ◽  
Older Adults ◽  
Natural History ◽  
Confidence Interval ◽  
Odds Ratio ◽  
Death Rate ◽  
Mortality Data ◽  
Total Death ◽  
History Of

The purpose of the study was to investigate the natural history of insomnia and its association with depression and mortality. In 1983, 1,870 randomly selected subjects aged 45–65 years answered a questionnaire on sleep and health. Of the 1,604 survivors in 1995, 1,244 (77.6%) answered a new questionnaire with almost identical questions. Mortality data were collected for the 266 subjects that had died during the follow-up period. Chronic insomnia was reported by 36.0% of women and 25.4% of men (χ2 = 9.7; p < .01). About 75% of subjects with insomnia at baseline continued to have insomnia at follow-up. Insomnia in women predicted subsequent depression (odds ratio [OR] = 4.1; 95% confidence interval [CI] 2.1–7.2) but was not related to mortality. In men, insomnia predicted mortality (OR = 1.7; 95% CI 1.2–2.3), but after adjustment for an array of possible risk factors, this association was no longer significant. Men with depression at baseline had an adjusted total death rate that was 1.9 times higher than in the nondepressed men (95% CI: 1.2–3.0).

Sign in / Sign up

Export Citation Format

Share Document