Biostatistical Associations and Machine-Learning Classification of Measured Blood Pressure Threshold Status Among U.S. Adults: NHANES August 2021–August 2023

Q: What are the key findings of this study?

Age, male sex (OR 1.5258, p<0.001), BMI, and smoking were associated with blood pressure ≥130/80 mmHg, while machine learning models showed modest discrimination (test AUCs 0.654–0.669).

Q: What is the clinical evidence from this study?

Study design: Cross-Sectional. Population: Measured blood pressure threshold status (systolic ≥130 mmHg or diastolic ≥80 mmHg) (n=6123). Intervention: Male sex (alongside age, body mass index, and smoking history in multivariable model; machine learning classification using logistic regression, random forest, and SVM) vs. Female sex (reference group). Primary outcome: Mean systolic blood pressure ≥130 mmHg or mean diastolic blood pressure ≥80 mmHg during a single NHANES examination (OR 1.5258, 95% CI 1.2827-1.8150, p=<0.001).

Q: What does this research mean for the field?

Machine-learning classifiers utilizing demographic and lifestyle factors achieve only moderate discrimination (test AUCs 0.654–0.669) for identifying measured blood pressure ≥130/80 mmHg in U.S. adults. Novelty: ClaimNovelty.INCREMENTAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

Q: What question did this study set out to answer?

To examine the biostatistical associations of demographic and lifestyle factors with measured blood pressure status (systolic ≥130 mmHg or diastolic ≥80 mmHg) and evaluate three machine-learning classifiers.

Zinah Arkah

doi:10.36371/port.2026.4.36

October 31, 2026Journal Port Science ResearchOpen Access

Biostatistical Associations and Machine-Learning Classification of Measured Blood Pressure Threshold Status Among U.S. Adults: NHANES August 2021–August 2023

View Full Paper

Ask AI

Bookmark

Share

Key result

Male sex linked to ~53% higher odds of blood pressure ≥130/80 mmHg.

OR 1.5258
95% CI 1.2827-1.8150
P<0.001
n=6,123

Why the study?

The study examined associations of demographic and lifestyle characteristics with elevated measured blood pressure among U.S. adults and evaluated machine-learning models for classifying blood pressure threshold status.

Do demographic and lifestyle characteristics (age, sex, BMI, smoking history, sedentary time, and leisure-time physical activity) associate with and accurately classify measured blood pressure ≥130/80 mmHg among U.S. adults in NHANES 2021–2023?

Key Points

To examine the biostatistical associations of demographic and lifestyle factors with measured blood pressure status (systolic ≥130 mmHg or diastolic ≥80 mmHg) and evaluate three machine-learning classifiers.
Analyzed NHANES cross-sectional data (August 2021–August 2023) from 6,123 adults with valid blood pressure readings.
Performed survey-adjusted logistic regression on 5,985 complete cases to estimate adjusted odds ratios.
Evaluated logistic regression, random forest, and support vector machine models using a stratified training–test split and five-fold cross-validation on the training set.
Meeting the threshold was significantly associated with age (OR = 1.0325, 95% CI: 1.0283–1.0368; p < 0.001), male sex (OR = 1.5258, 95% CI: 1.2827–1.8150; p < 0.001), BMI (OR = 1.0594, 95% CI: 1.0457–1.0732; p < 0.001), and lifetime smoking history (OR = 1.1400, 95% CI: 1.0089–1.2881; p = 0.038).
Model discrimination was moderate on the independent test set, with test AUCs of 0.6677 for logistic regression, 0.6686 for random forest, and 0.6539 for support vector machines.
Lowering the logistic regression decision threshold from 0.50 to 0.31 increased recall from 0.3428 to 0.8560 while specificity decreased from 0.7828 to 0.4317.

Study Design

Type

Cross-Sectional (n=6,123)

Structured PICO

Do demographic and lifestyle characteristics (age, sex, BMI, smoking history, sedentary time, and leisure-time physical activity) associate with and accurately classify measured blood pressure ≥130/80 mmHg among U.S. adults in NHANES 2021–2023?

P

Population

6,123 U.S. adults aged 18–80 years (mean age 52.42 years) from the NHANES August 2021–August 2023 cycle with at least one valid systolic and diastolic blood pressure measurement.

E

Exposure

Survey-adjusted multivariable logistic regression and machine-learning classification models (logistic regression, random forest, and support vector machine) using a six-variable demographic and lifestyle predictor set (age, sex, body mass index, sedentary time, lifetime smoking history of ≥100 cigarettes, and binary leisure-time physical activity indicator), alongside classification threshold tuning (0.31 vs 0.50).

O

Outcome

Measured blood pressure threshold status from a single NHANES examination, defined as mean systolic blood pressure ≥130 mmHg or mean diastolic blood pressure ≥80 mmHg.surrogate

Main Result

Odds Ratio: 1.5258 (95% CI 1.2827–1.815)

p-value: p=<0.001

Limitations

Cross-sectional data represent simultaneous measurements rather than predicting future hypertension.
Single-visit blood pressure measurement (≥130/80 mmHg) is not a confirmed clinical diagnosis and ignores blood pressure controlled by medication.
Predictors were constrained, and smoking history and physical activity were self-reported using simplified indicators.
Machine-learning metrics did not use survey weighting, the train-test split did not respect survey clusters, and results lacked external validation.
Confidence intervals for predictive performance metrics, model calibration, and decision-curve net benefit were not assessed.
Cross-sectional design captures simultaneous measurements rather than predicting future hypertension.
The single-visit 130/80 mmHg threshold classification is not a confirmed clinical diagnosis and does not account for blood pressure controlled by medication.
Constrained predictor set relying on self-reported smoking history (which does not distinguish current from former smokers) and a binary leisure-time physical activity indicator that does not reflect total activity volume.
Machine-learning metrics did not incorporate NHANES survey weights or clusters and were evaluated on a single train-test split without external validation.
Did not report confidence intervals for predictive performance metrics or assess model calibration or decision-curve net benefit.

Implication

Although age, male sex, BMI, and lifetime smoking history are significantly associated with blood pressure ≥130/80 mmHg at the population level in NHANES 2021–2023, machine-learning models using only basic demographic and lifestyle predictors show only modest individual discrimination (test AUC 0.654–0.669) and are insufficient for standalone clinical screening without further development and external validation.

Abstract

This cross-sectional analysis utilized publicly available NHANES data collected from August 2021-2023. The study included 6,123 adults for whom at least one valid systolic and one valid diastolic blood pressure reading was available.We examined the association of demographic and lifestyle characteristics with blood pressure status, and evaluated three machine-learning classifiers. Participants were classified according to their mean measurements for the objective of this analysis: systolic blood pressure ≥130 mmHg or diastolic blood pressure ≥80 mmHg. Survey-adjusted logistic regression was based on 5,985 complete cases. We evaluated logistic regression, random forest, and support vector machine models using a stratified training–test split and five-fold cross-validation on the training data. Age, male sex, body mass index, and lifetime cigarette-smoking history were associated with the outcome. Test AUCs ranged from 0.654 to 0.669. Lowering the logistic-regression threshold from 0.50 to 0.31 increased recall from 0.343 to 0.856, while specificity fell from 0.783 to 0.432. The level of discrimination was moderate, and an external validation process is required before any clinical application.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Zinah Arkah (2026) conducted a cross-sectional in Measured blood pressure threshold status (systolic ≥130 mmHg or diastolic ≥80 mmHg) (n=6,123). Male sex (alongside age, body mass index, and smoking history in multivariable model; machine learning classification using logistic regression, random forest, and SVM) vs. Female sex (reference group) was evaluated on Mean systolic blood pressure ≥130 mmHg or mean diastolic blood pressure ≥80 mmHg during a single NHANES examination (OR 1.5258, 95% CI 1.2827-1.8150, p=<0.001). Age, male sex (OR 1.5258, p<0.001), BMI, and smoking were associated with blood pressure ≥130/80 mmHg, while machine learning models showed modest discrimination (test AUCs 0.654–0.669).

synapsesocial.com/papers/6ae7d421edffb4962941048e https://doi.org/10.36371/port.2026.4.36

View Full Paper

Ask AI

Bookmark

Share

Biostatistical Associations and Machine-Learning Classification of Measured Blood Pressure Threshold Status Among U.S. Adults: NHANES August 2021–August 2023

Key Points

Study Design

Structured PICO

Main Result

Limitations

Implication

Abstract

Discussion

Cite This Study

Biostatistical Associations and Machine-Learning Classification of Measured Blood Pressure Threshold Status Among U.S. Adults: NHANES August 2021–August 2023

Key Points

Study Design

Structured PICO

Main Result

Limitations

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading