A Machine Learning Classification Model for Gastrointestinal Health in Cancer Survivors: Roles of Telomere Length and Social Determinants of Health DOI Open Access
Claire J. Han, Xia Ning, Christin E. Burd

et al.

International Journal of Environmental Research and Public Health, Journal Year: 2024, Volume and Issue: 21(12), P. 1694 - 1694

Published: Dec. 19, 2024

Background: Gastrointestinal (GI) distress is prevalent and often persistent among cancer survivors, impacting their quality of life, nutrition, daily function, mortality. GI health screening crucial for preventing managing this distress. However, accurate classification methods remain unexplored. We aimed to develop machine learning (ML) models classify status (better vs. worse) by incorporating biological aging social determinants (SDOH) indicators in survivors. Methods: included 645 adult survivors from the 1999–2002 NHANES survey. Using training test datasets, we employed six ML conditions worse). These incorporated leukocyte telomere length (TL), SDOH, demographic/clinical data. Results: Among models, random forest (RF) performed best, achieving a high area under curve (AUC = 0.98) dataset. The gradient boosting (GBM) demonstrated excellent performance with AUC (0.80) TL, several socio-economic factors, risk behaviors (including lifestyle choices), inflammatory markers were associated health. most significant input features better our longer TL an annual household income above poverty level, followed routine physical activity, low white blood cell counts, food security. Conclusions: Our findings provide valuable insights into classifying identifying factors related health, including SDOH indicators. To enhance model predictability, further longitudinal studies external clinical validations are necessary.

Language: Английский

A Measure of Nutrition Security Using the National Health and Nutrition Examination Survey Dataset DOI Creative Commons
Vibha Bhargava, Jung Sun Lee, Travis A. Smith

et al.

JAMA Network Open, Journal Year: 2025, Volume and Issue: 8(2), P. e2462130 - e2462130

Published: Feb. 28, 2025

Accurate and practical measures of nutrition security are needed to assess monitor its prevalence identify associated risk factors in the US. To propose a measure derived from combining self-assessed food diet quality indicators available National Health Nutrition Examination Survey (NHANES) sociodemographic health security. This cross-sectional study used data 6 cycles NHANES, which collects on general status behaviors, dietary intake, physiological measurements, characteristics, conducted 2007 2018. All participants were adults aged 20 years or older. Statistical analysis was performed between October 2023 April 2024. Sociodemographic including age, sex, race ethnicity, marital status, household size, presence children household, educational level, poverty income ratio (PIR), Supplemental Assistance Program (SNAP) participation, weight chronic conditions, insurance coverage. A security, measured using US Department Agriculture Household Food Security Module, self-rated indicators. Four categories created dichotomized measures: secure with high (FSHD), low (FSLD), insecure (FIHD), (FILD). Only respondents classified as FSHD considered be secure. The unweighted analytic sample included 28 898 NHANES participants. weighted mean [SD] age 47.3 [14.5] years; 51.9% (weighted) female; 11.1% identified Black, 13.6% Hispanic, 67.4% White individuals; 35.6% those surveyed by proposed (ie, FSLD, FIHD, FILD). Of these participants, 20.2% (95% CI, 19.4%-21.0%) categorized being due 8.4% 7.8%-9.1%) 7.0% 6.4%-7.6%) FILD. remaining 64.4% 63.2%-65.7%) secure). Younger (20-44 years: average marginal effect [AME], -0.193; 95% -0.217 -0.168), (PIR <1.30: AME, -0.111; -0.136 -0.085), lower level (≤high school diploma: -0.135; -0.156 -0.114), racial ethnic minority (Hispanic: -0.054; -0.075 -0.032), SNAP participation (AME, -0.073; -0.099 -0.047), obesity -0.118; -0.138 -0.097), self-reported fair poor -0.239; -0.260 -0.217) insecurity. feasible for assessing monitoring validated NHANES. laid groundwork exploring other national datasets performing regular collection key dimensions assessment

Language: Английский

Citations

1

Food insecurity and risk of nutrition insecurity among supplemental nutrition assistance program participants in Rhode Island and connecticut, USA DOI Creative Commons
Vanessa M. Oddo, Julien Leider, Alison Tovar

et al.

Preventive Medicine Reports, Journal Year: 2025, Volume and Issue: 51, P. 103002 - 103002

Published: Feb. 8, 2025

To 1) describe food insecurity and risk of nutrition (henceforth insecurity); 2) test the associations between perceived access neighborhood environment insecurity, differences in these associations; 3) diet quality, among a sample adults with low income. Between May-September 2023, Supplemental Nutrition Assistance Program (SNAP) participants Rhode Island Connecticut, USA (n = 1234) completed frequency questionnaire, from which we calculated healthy eating index (HEI)-2015 scores. An online survey included questions on barriers environment, security. We used separate adjusted regression models to estimate correlates associated and/or their quality. Individuals were 35 years old, average, 92 % women, 43 identified as Hispanic, 58 30 insecure, respectively. The average HEI-2015 score was 64. Lack money an 8-fold higher odds experiencing (95 Confidence Interval [CI] 5.76, 10.67). largest magnitude association having few or no full-service grocery stores nearby (Odds Ratio[OR] 2.27; 95 CI 1.27, 4.06), followed by lack limited transportation. Associations negative but not statistically significant. prevalence SNAP than Americans, average. Perceived insecurity.

Language: Английский

Citations

0

A New Approach To Guide Research And Policy At The Intersection Of Income, Food, Nutrition, And Health DOI
Seth A. Berkowitz, Hilary K. Seligman,

Dariush Mozaffarian

et al.

Health Affairs, Journal Year: 2025, Volume and Issue: 44(4), P. 384 - 390

Published: April 1, 2025

Income distribution, food and nutrition insecurity, poor diet quality contribute to diet-related disease, which is a major threat population health equity. Based on our review synthesis of the empirical evidence, we provide new conceptual model for understanding interrelationships among income, security, quality, health. We identify directions future research discuss policy program implications model. Overall, interventions that address income security can facilitate, but do not ensure, better although they improve in other ways. Importantly, even people who are secure have adequate frequently unhealthy diets. Addressing these challenges will require innovative policies Such should include efforts increase availability accessibility Food Is Medicine care. Health insurance coverage evidence-based, clinically indicated programs critical success efforts.

Language: Английский

Citations

0

Construct Validity of Nutrition Security and Perceived Healthfulness Choice Measures DOI
Patrick Brady, Melissa N. Laska

Journal of Hunger & Environmental Nutrition, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 11

Published: April 28, 2025

Language: Английский

Citations

0

Developing methods and measures for assessing and monitoring nutrition security DOI
Edward A. Frongillo

American Journal of Clinical Nutrition, Journal Year: 2024, Volume and Issue: 119(6), P. 1381 - 1382

Published: May 13, 2024

Language: Английский

Citations

0

An evidence-based definition of nutrition security: disparities in sociodemographic characteristics, dietary intake and cardiometabolic risk using the US Healthy Eating Index DOI Creative Commons

Elise Sheinberg,

Laura A. Schmidt,

Jerold R. Mande

et al.

medRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: June 4, 2024

ABSTRACT Importance Establishing a universal metric for nutrition security, defined as, “consistent and equitable access to healthy, safe, affordable foods essential optimal health well-being,” is national priority. Understanding how the Healthy Eating Index-2020 (HEI-2020) could be used measure monitor security in population can assist surveillance improve design of programs policies. Objective To examine utility HEI-2020 as an evidence-based security. Design This serial cross-sectional study data from 2009-2018 National Health Nutrition Examination Surveys (NHANES). Setting Nationally representative, population-based survey Participants Data N=22,168 US adults (≥20 years) were analyzed. Main outcome We derived scores (0-100), commonly capture overall diet quality, participants’ two 24-hour dietary recalls. categories established: high (>70-100), marginal (>60-70), low (>50-60), very (0-50). Results Only 13% had while nearly two-thirds or was higher who older, female, “Other” race ethnicity, born outside US, have education attainment, income, food Compared with lowest intakes unprocessed minimally processed foods, fruits, vegetables, whole grains, seafood highest ultra-processed refined red meats (all P-trends <0.01 ). Similarly, more likely elevated adiposity, lower HDL cholesterol, triglycerides, fasting glucose, hemoglobin A1c ≤0.01 Conclusions The HEI robust that directly linked construct Using cut-points would allow policy makers, public practitioners professionals set benchmarks nationwide targets achieving KEY POINTS Question Can Index (HEI)-2020 security? Findings nationally representative data, we created four using scores: (≥70), (>60-70, (≤50). Less than 1 6 High less prevalent among greater socioeconomic disadvantage. also associated favorable cardiometabolic risk profiles. Meaning program setting.

Language: Английский

Citations

0

A Machine Learning Classification Model for Gastrointestinal Health in Cancer Survivors: Roles of Telomere Length and Social Determinants of Health DOI Open Access
Claire J. Han, Xia Ning, Christin E. Burd

et al.

International Journal of Environmental Research and Public Health, Journal Year: 2024, Volume and Issue: 21(12), P. 1694 - 1694

Published: Dec. 19, 2024

Background: Gastrointestinal (GI) distress is prevalent and often persistent among cancer survivors, impacting their quality of life, nutrition, daily function, mortality. GI health screening crucial for preventing managing this distress. However, accurate classification methods remain unexplored. We aimed to develop machine learning (ML) models classify status (better vs. worse) by incorporating biological aging social determinants (SDOH) indicators in survivors. Methods: included 645 adult survivors from the 1999–2002 NHANES survey. Using training test datasets, we employed six ML conditions worse). These incorporated leukocyte telomere length (TL), SDOH, demographic/clinical data. Results: Among models, random forest (RF) performed best, achieving a high area under curve (AUC = 0.98) dataset. The gradient boosting (GBM) demonstrated excellent performance with AUC (0.80) TL, several socio-economic factors, risk behaviors (including lifestyle choices), inflammatory markers were associated health. most significant input features better our longer TL an annual household income above poverty level, followed routine physical activity, low white blood cell counts, food security. Conclusions: Our findings provide valuable insights into classifying identifying factors related health, including SDOH indicators. To enhance model predictability, further longitudinal studies external clinical validations are necessary.

Language: Английский

Citations

0