Search Results (68)

Social determinants of health (SDoH) comprise a broad array of social conditions, such as access to food and housing, that facilitate or impede individual behavior. The aim of this study was to assess the association between SDoH and cigarette smoking-related outcomes among U.S. adults (aged ≥18 years) by using data from the 2023 Behavioral Risk Factor Surveillance System (BRFSS). Cross-sectional data were obtained from the Social Determinants and Health Equity (SD/HE) module, conducted in 33 states, the District of Columbia, and Puerto Rico as part of the 2023 BRFSS. We examined four indicators of adverse SDoH (i.e., food insecurity, housing insecurity, utility insecurity, and lack of reliable transportation) and three cigarette smoking-related outcomes (i.e., cigarette smoking status, menthol cigarette smoking, and past-year quit attempt). All analyses were conducted with SAS 9.4 and used BRFSS sampling weights to adjust for the complex sampling design. Among 45,160 respondents, 2991 (7.8%) were adults who smoked cigarettes in the past month, of whom 570 (16.5%) reported making a quit attempt in the past 12 months. Menthol cigarette use was reported by 634 (22.0%) adults who smoked cigarettes in the past month. In adjusted analyses, each SDoH measure (i.e., food insecurity (aOR = 1.70, 95% CI: 1.19–2.41, p < 0.01), housing insecurity (aOR = 1.66, 95% CI: 1.06–2.59, p < 0.05), utility insecurity (aOR = 1.92, 95% CI: 1.01–3.65, p < 0.05), and lack of reliable transportation (aOR = 1.67, 95% CI: 1.03–2.73, p < 0.05)) was significantly associated with making a quit attempt in the past 12 months. Food insecurity was significantly associated with the odds of current cigarette smoking. Food insecurity and utility insecurity were independent risk factors for using menthol cigarettes. U.S. adults experiencing adverse SDoH are trying to stop smoking at higher rates than adults not experiencing adverse SDoH. Findings demonstrate that SDoH is a strong predictor of cigarette smoking status, menthol cigarette smoking, and past-year quit attempts among U.S. adults. Full article

Background: The literature shows a strong association between level of education and the risk of developing Coronary Heart Disease (CHD). However, the extent to which this association attenuates after accounting for sociodemographic characteristics and cardiovascular risk factors in a survey-weighted national sample warrants further evaluation. Objective: We aimed to assess the association between educational attainment and angina and myocardial infarction (MI) across age groups in a nationally representative U.S. sample. Methods: The study analyzed 2021 Behavioral Risk Factor Surveillance System (BRFSS) data from 438,693 adults, a nationally representative telephone survey of U.S. adults. The dataset was accessed from the Centers for Disease Control and Prevention BRFSS website in February 2023. Angina and MI were identified based on self-reported physician diagnoses. Analyses included adults aged 18 years and older with no missing data for education and outcomes. Results: In survey-weighted analyses with college graduates as the reference group, lower educational attainment was associated with higher odds of angina and MI, compared with college graduates. In the fully adjusted model (Model 2), attending high school was associated with higher odds of angina (OR 1.439) and MI (OR 2.390). Conclusions: Lower educational attainment is associated with higher odds of angina and MI, particularly among younger adults. Although the magnitude of these associations was attenuated after adjustment for sociodemographic and cardiovascular risk factors, the persistence of the association underscores the importance of considering educational disparities in cardiovascular risk assessment. Full article

Because CKMS was only proposed by the American Heart Association in 2023, there has been a paucity of information about the distribution and determinants of the syndrome across population groups. We reviewed published studies of the prevalence of CKMS in the U.S. and other countries and obtained new estimates of the prevalence of this syndrome among U.S. adults by birth decade and sociodemographic attributes using 2019, 2021, and 2023 Behavioral Risk Factor Surveillance System (BRFSS) data. The results of this study indicate that CKMS is widespread in the general U.S. population, especially among older cohorts born before 1940 and during the 1940s, 1950s, and 1960s. Except for the three younger cohorts, born in the 1980s, 1990s, and 2000 or later, the prevalence of CKMS stage 4 was significantly higher among males than in females. Among those born between the 1950s and 1990s, the prevalence was significantly higher among non-Hispanic Blacks compared to their non-Hispanic white counterparts. Across all birth decades, prevalence of CKMS stage 4 was generally higher among those without a college degree, from a low-income household, and residing in rural areas. These prevalence rate estimates will further our understanding of the burden and unique needs of different population groups in improving cardiovascular–kidney–metabolic health across the life course. Full article

Machine learning is increasingly applied to cardiovascular disease prediction yet reported performance metrics often appear implausibly high due to methodological errors. Recent work has reported nearly perfect predictive accuracy (≈99%) using a k-Nearest Neighbors (kNN) model on CDC heart-disease data. Such performance greatly exceeds typical BRFSS-based benchmarks and strongly indicates data leakage. In this commentary, we replicate and re-analyze the original workflow, showing that the authors applied the SMOTE-ENN resampling method prior to the train/test split, thereby allowing synthetic data generated from the full dataset to contaminate the test set. Combined with an excessively small neighborhood parameter (k = 2), this produced misleadingly high accuracy. It is noted that (1) with SMOTE-ENN performed globally, synthetic samples appear nearly identical to test points, leading to near-perfect classification, and (2) this kNN choice is unusually small for a dataset of this scale and further amplifies leakage bias. Correcting the workflow by restricting oversampling to the training data or using undersampling restores realistic results, reducing predictive accuracy to approximately 80%, confirming the inflation caused by pre-split resampling and aligning with literature norms. This case underscores the critical importance of rigorous validation, transparent reporting, and leakage-free pipelines in medical AI. We outline practical guidelines for avoiding such pitfalls and ensuring reproducible, realistic, and clinically reliable machine-learning studies. Full article

Objective: To evaluate population-level temporal relationships between modifiable lifestyle factors and rising breast, colorectal and uterine cancer incidence rates among females under 50 years old. Methods: This retrospective ecological study utilized data from the United States Cancer Statistics (USCS) for cancer incidence, the National Health and Nutrition Examination Survey (NHANES) for health-related behaviors, and the Behavioral Risk Factor Surveillance System (BRFSS) for physical activity. Modifiable lifestyle factors analyzed included obesity (BMI ≥ 30 kg/m²), smoking, alcohol use, fiber and saturated fat intake, caloric intake, and physical activity. Trends were assessed using average annual percent change (AAPC), and population-level correlations between cancer incidence and lifestyle factors were evaluated using Pearson correlation coefficients. Results: Between 2001 and 2018, 914,659 breast, 144,130 colorectal, and 124,399 uterine cancer cases were identified. The largest increases in cancer incidence occurred in age groups under 30 years old. Colorectal cancer increased by 6.9%, followed by uterine cancer at 4.8% and breast cancer at 1.7%, all p < 0.001. When examining this age group by race, colorectal cancer increased by 8.0% (p < 0.001) annually in White women aged 20–24 years, while uterine cancer rose 4.8% (p < 0.001) in Hispanic women in the 20–24 and 25–29 year age groups. Breast cancer also increased by 2.0% (p < 0.001) per year in White women 25–29 years old. Smoking rates decreased, and alcohol consumption and obesity rates increased. No significant correlation was found between cancer incidence and smoking, caloric intake, saturated fat, or physical activity. A moderate positive correlation was identified between alcohol use and cancer risk (r = 0.55–0.67, p < 0.05). Obesity prevalence showed strong population-level temporal correlation with cancer incidence for all three cancers with stratified analysis demonstrating the strongest correlations in patients with class III obesity. Conclusions: From 2001 to 2018, the incidence of breast, colorectal, and uterine cancers increased most sharply among women under 30 years of age. Over the same period, obesity prevalence in this population also increased. These population-level observations are hypothesis-generating and require confirmation in individual-level, prospective studies to determine whether and how obesity and other lifestyle factors influence early-onset cancer risk. Full article

Cancer is the second leading cause of death in the U.S., with 612,000 deaths estimated in 2023. Cancer screening (CS) reduces mortality through early detection, but the impact of chronic conditions like adult asthma (AA) on screening is less understood. This study explores the association between AA and uptake of prostate, breast, cervical, and colorectal CS using BRFSS 2016, 2018, 2020 and 2022 data. Weighted and adjusted multivariable logistic regression assessed the association between AA and CS across sex and age-based subgroups with significant effect modification testing and subsequent subgroup analyses. Results showed significantly higher CS adherence among individuals with AA across all four CS sites compared to counterparts without asthma (CCWA). Males (55–69 years old (YO)) with AA had 15% significantly higher weighted and adjusted odds (WAO) of prostate CS (95% CI: 1.04–1.27). Women (50–74 YO) with AA had 16% significantly higher WAO of breast CS (95% CI: 1.01–1.32), with non-depressed, heavy-drinking women showing 300% significantly higher WAO (95% CI: 2.20–7.22) of breast CS. Women (21–65 YO) with AA had 9% significantly higher WAO of cervical CS (95% CI: 1.02–1.17), with education significantly modifying the association (WAOR for college-educated women = 1.23, 95% CI: 1.11–1.36). When CCWA, colorectal CS showed significantly higher odds of 36% for men aged 50–75 (95% CI: 1.24–1.49) and 24% for women aged 50–75 (95% CI: 1.15–1.33). This is the first national study to examine the association between AA and uptake of prostate, female breast, cervical, and colorectal CS over four years. Individuals with AA had significantly greater odds of CS adherence than CCWA. Effect modification by heavy drinking and education suggests the need for targeted outreach and low-literacy interventions. Full article

Background: Diabetes disproportionately affects U.S. subgroups, yet non-Hispanic multiracial adults are underrepresented in epidemiologic studies. This study aimed to examine behavioral and sociodemographic predictors of diabetes in this population. Methods: We analyzed data from the 2023 Behavioral Risk Factor Surveillance System (BRFSS) using a cross-sectional design that incorporated survey weights, strata, and primary sampling units. Binary logistic regression was employed to identify predictors of diabetes, including variables with p ≤ 0.20 from bivariate models in the multivariable analysis. Adjusted odds ratios (AORs) with 95% confidence intervals (CIs) were reported. Results: The study included a total of 6429 participants. Obesity (AOR = 4.16; 95% CI: 3.33, 33.23), being overweight (AOR = 2.05; 1.62, 2.60), poor general health (AOR = 2.82; 2.38, 38.35), age ≥ 65 years (AOR = 3.08; 2.60, 3.65), male sex (AOR = 1.34; 1.15, 1.58), and health insurance (AOR = 2.14; 1.35, 3.61) were associated with higher odds of diabetes. Physical activity (AOR = 0.76; 0.64, 0.90) and alcohol consumption (AOR = 0.55; 0.47, 47.65) were linked to lower odds of diabetes. Smoking status showed no significant association after adjustment. Conclusions: In non-Hispanic multiracial adults, factors such as adiposity and older age increased the risk of diabetes, while physical activity and alcohol consumption offered protective benefits. These findings indicate that current diabetes prevention strategies are applicable to this subgroup, and public health initiatives should prioritize their inclusion in outreach, screening, and intervention efforts. Full article

Background: Chronic diseases are a significant and escalating public health concern in the United States (U.S.) and globally. Chronic co-morbidities such as heart disease, stroke, diabetes, other cardiovascular diseases, and asthma are major risk factors for death and disability. Behavioral factors such as smoking, alcohol use, sedentary lifestyle, and poor dietary habits are among the major risk factors leading to these chronic diseases. Purpose: This study aims to investigate how combinations of unhealthy behaviors are associated with the risk of cardiovascular diseases in various populations. Methods: Using data from the 2023 Behavioral Risk Factor Surveillance System (BRFSS), we computed multivariable logistic regression models to assess the association of unhealthy behaviors with the risk of chronic diseases. Results: Our results show that compounded score of risky health behaviors such as smoking, alcohol use, and physical inactivity, as well as other covariates such as older age, being male, previously married, living in a rented house, unemployed, living in non-metropolitan counties, having high blood pressure, and high cholesterol, were associated with experiencing a heart attack, coronary heart disease, and stroke. Conclusions: Our results highlight the need for behavior-focused population health interventions to lower morbidity and health inequities by showing that unhealthy behaviors and sociodemographic disparities significantly raise the risk of cardiovascular diseases. Full article

The use of Artificial Intelligence (AI) in healthcare has significantly advanced early disease detection, enabling timely diagnosis and improved patient outcomes. This work proposes an end-to-end machine learning (ML) model for predicting diabetes based on data quality by following key steps, including advanced preprocessing by KNN imputation, intelligent feature selection, class imbalance with a hybrid approach of SMOTEENN, and multi-model classification. We rigorously compared nine ML classifiers, namely ensemble approaches (Random Forest, CatBoost, XGBoost), Support Vector Machines (SVM), and Logistic Regression (LR) for the prediction of diabetes disease. We evaluated performance on specificity, accuracy, recall, precision, and F1-score to assess generalizability and robustness. We employed SHapley Additive exPlanations (SHAP) for explainability, ranking, and identifying the most influential clinical risk factors. SHAP analysis identified glucose levels as the dominant predictor, followed by BMI and age, providing clinically interpretable risk factors that align with established medical knowledge. Results indicate that ensemble models have the highest performance among the others, and CatBoost performed the best, which achieved an ROC-AUC of 0.972, an accuracy of 0.968, and an F1-score of 0.971. The model was successfully validated on two larger datasets (CDC BRFSS and a 130-hospital dataset), confirming its generalizability. This data-driven design provides a reproducible platform for applying useful and interpretable ML models in clinical practice as a primary application for future Internet-of-Things-based smart healthcare systems. Full article

Short sleep duration (≤6 h) is a public health concern linked to cardiometabolic disease and premature mortality. However, persistent disparities across sociodemographic, psychosocial, and structural domains remain underexplored in recent nationally representative samples. We analyzed 2022 Behavioral Risk Factor Surveillance System (BRFSS) data, including 228,463 adults (weighted N ≈ 122 million). Sleep duration was dichotomized as short (≤6 h) versus adequate (≥7 h). Complex samples logistic regression estimated associations between sociodemographic, psychosocial, behavioral, and structural determinants and short sleep, accounting for survey design. The weighted prevalence of short sleep was 33.2%. Non-Hispanic Black (AOR = 1.56, 95% CI: 1.46–1.65) and American Indian/Alaska Native adults (AOR = 1.46, 95% CI: 1.29–1.65) were disproportionately affected compared with non-Hispanic White adults. Psychosocial factors contributed strongly: life dissatisfaction, limited emotional support, and low social connectedness increased odds, whereas high connectedness was protective. Food insecurity and smoking were significant structural and behavioral risks, while binge drinking and urbanicity were not. One-third of U.S. adults report short sleep, with marked disparities across demographic, socioeconomic status, psychosocial stressors, and structural barriers. Findings highlight the multifactorial nature of sleep health inequities and the need for multilevel interventions addressing both individual behaviors and upstream determinants. Full article

Dental care is a well-established marker of both oral and systemic health, driven by behavioral, socioeconomic, and geographic factors. This study aimed to develop and evaluate machine learning models to classify the presence and severity of permanent tooth loss in U.S. adults using the 2022 Behavioral Risk Factor Surveillance System (BRFSS) dataset. We analyzed responses from 365,803 adults after recoding demographic, behavioral, socioeconomic, and access variables. Ten supervised classifiers were trained and evaluated using stratified 80/20 train–test splits, with ANOVA-based selection for the binary task and Pearson correlation plus engineered features for the multiclass task. Performance was assessed by accuracy, AUC, precision, recall, and specificity. For binary classification (any loss vs. none), XGBoost achieved the highest performance (AUC = 0.786; accuracy = 71.4%), with CatBoost close behind (AUC = 0.711). For multiclass severity (none, 1–5, 6+, all teeth removed), an ensemble of gradient-boosting models achieved strong discrimination (macro-AUC = 0.752). Key predictors included age, smoking, education, income, and general health. These findings demonstrate that large-scale survey–based ML models can support oral health surveillance by identifying high-risk groups and informing targeted prevention strategies. Full article

Despite growing recognition of the role that social determinants of health (SDOHs) and health-related social needs (HRSNs) play in chronic disease, limited research has examined their associations with Chronic Obstructive Pulmonary Disease (COPD) in population-based studies. This cross-sectional study analyzed 2022 Behavioral Risk Factor Surveillance System (BRFSS) data from 37 U.S. states and territories to determine how financial hardship, food insecurity, employment loss, healthcare access barriers, and psychosocial stressors influence the prevalence of COPD. Weighted logistic regression models were used to assess the associations between COPD and specific SDOHs and HRSNs. Several individual SDOH and HRSN factors were significantly associated with COPD prevalence, with financial strain emerging as a particularly strong predictor. In models examining specific SDOH factors, economic hardships like inability to afford medical care were strongly linked to higher COPD odds. Psychosocial HRSN risks, such as experiencing mental stress, also showed moderate associations with increased COPD prevalence. These findings suggest that addressing both structural and individual-level social risks may be critical for reducing the prevalence of COPD in populations experiencing financial challenges. Full article

Large language models (LLMs) are increasingly being applied to clinical tasks, but it remains unclear whether medical-specific models consistently outperform smaller, generalpurpose ones. This study investigates that assumption in the context of cardiovascular disease (CVD) risk assessment. We fine-tuned eight LLMs—both general-purpose and medical-specific—using textualized data from the Behavioral Risk Factor Surveillance System (BRFSS) to classify individuals as “High Risk” or “Low Risk”. To provide actionable insights, we integrated a Retrieval-Augmented Generation (RAG) framework for personalized recommendation generation and deployed the system within an interactive chatbot interface. Notably, Gemma2, a compact 2B-parameter general-purpose model, achieved a high recall (0.907) and F1-score (0.770), performing on par with larger or medical-specialized models such as Med42 and BioBERT. These findings challenge the common assumption that larger or specialized models always yield superior results, and highlight the potential of lightweight, efficiently fine-tuned LLMs for clinical decision support—especially in resource-constrained settings. Overall, our results demonstrate that general-purpose models, when fine-tuned appropriately, can offer interpretable, high-performing, and accessible solutions for CVD risk assessment and personalized healthcare delivery. Full article

Mississippi has long been one of the most obese states in the U.S., with its obesity rates consistently exceeding the national average. The state’s severe obesity rate is also among the highest in the nation. This study utilized the 2011 to 2021 data from the Mississippi Behavioral Risk Factor Surveillance System (BRFSS) to conduct a comprehensive analysis of obesity and severe obesity trends in Mississippi by sex, age, and race and ethnicity. The data set included a BMI variable calculated by using self-reported height and weight, which the authors categorized into two obesity classification groups—obesity (BMI: 30.00 to 39.99) and severe obesity (BMI: 40.00 or greater)—and demographic characteristics such as sex, age, race and ethnicity. The data were analyzed using SAS 9.4 software to account for the complex design. Weighted prevalence estimates and associated standard errors (SEs) for obesity and severe obesity were calculated. Changes in the prevalence over time were assessed using logistic regression models. The prevalence estimates and SEs were exported to Joinpoint software to calculate the annual percentage change (APC) and associated 95% confidence intervals (CIs) and p-values for the trends. Our analysis of the data revealed a consistent increase in severe obesity, regardless of age, sex, or race. A concerning trend exists where individuals are moving from the obese category to the severely obese category, indicating a worsening trend in overall weight status. This is likely to accelerate the development of chronic disease and, hence, place additional strain on an economically disadvantaged state. Future research should explore the underlying drivers of this shift, including biological, behavioral, and socioeconomic factors, while also evaluating the effectiveness of existing obesity prevention and treatment programs. Full article

This study aims to examine the mediating effects of health literacy on the relationship between frailty and medical costs among community-dwelling older adults. Methods: This study conducted a secondary data analysis of the research data that were constructed by linking the Korean Frailty and Aging Cohort Data (KFACD) and the National Health Insurance Database (NHID). Frailty was measured using the Modified Fried Phenotype. Medical costs were calculated using insurance-covered medical costs, including both inpatient and outpatient medical costs, from January 1 to December 31 of the year when the participants were enrolled in the Korean Frailty and Aging Cohort Study. Health literacy was assessed using questions from the Behavioral Risk Factor Surveillance System (BRFSS) conducted by the US Centers for Disease Control and Prevention. To examine the mediating role of health literacy in the relationship between frailty and medical costs, Baron and Kenny’s method was used. Linear regression was applied to estimate the association between frailty and health literacy, and Poisson regression was used to model the relationship between frailty, health literacy, and medical costs. Results: Frailty showed a negative correlation with health literacy (r = −0.27, p < 0.001) and a positive correlation with medical costs (r = 0.15, p < 0.001). Health literacy had a negative correlation with medical costs (r = −0.07, p = 0.008). We verified that health literacy played a partial mediating role in the relationship between frailty and medical costs. Conclusions: To reduce medical costs in older adults, intervention measures to improve health literacy as well as prevention and management measures for frailty should be considered simultaneously. In addition, primary medical institutions’ active participation in such projects is needed. Full article