A novel nutritional index for hypertension risk and prediction: Cross-sectional, prospective, and trajectory evidence from the English Longitudinal Study of Ageing

Abstract

Objective

Hypertension is a common and modifiable cardiovascular risk factor. The triglyceride–total cholesterol–body weight index is a novel metabolic indicator derived from routine clinical measurements, but its association with hypertension remains unclear.

Methods

This longitudinal cohort study used data from 5,731 middle-aged and older adults from the English Longitudinal Study of Ageing, with cross-sectional, prospective, and trajectory analyses. Hypertension was defined as a mean systolic blood pressure ≥140 mmHg, a mean diastolic blood pressure ≥90 mmHg, antihypertensive medication use, or self-reported physician-diagnosed hypertension. For the prospective analysis, participants without hypertension at baseline were followed for a median of 12 years. For the trajectory analysis, repeated triglyceride–total cholesterol–body weight index measurements from previous waves were used to identify long-term triglyceride–total cholesterol–body weight index patterns and assess their associations with hypertension risk. Subgroup analyses and propensity score matching were performed to assess the robustness of the findings.

Results

In the cross-sectional analyses, each one-unit increase in log-transformed triglyceride–total cholesterol–body weight index was associated with 43% higher odds of hypertension (adjusted odds ratio = 1.43, 95% confidence interval: 1.31–1.55). In the prospective analyses, baseline triglyceride–total cholesterol–body weight index predicted incident hypertension (adjusted hazard ratio = 1.16, 95% confidence interval: 1.05–1.28). Trajectory analyses identified low, moderate, and high triglyceride–total cholesterol–body weight index patterns, showing a graded increase in hypertension risk. Compared with the low-trajectory group, the high-triglyceride–total cholesterol–body weight index group had 69% higher odds of hypertension (adjusted odds ratio = 1.69, 95% confidence interval: 1.37–2.09, p < 0.001).

Conclusion

Triglyceride–total cholesterol–body weight index was independently and positively associated with the risk of hypertension and may provide a simple measure of metabolic burden relevant to early risk stratification.

Introduction

Hypertension is one of the most common and modifiable cardiovascular risk factors worldwide, contributing substantially to global morbidity and mortality. More than 1.3 billion adults are currently affected, and the prevalence continues to rise. ¹ Despite continuous advances in pharmacological therapy and prevention strategies, awareness and control rates of hypertension remain suboptimal in most regions. ² Increasing evidence indicates that metabolic disturbances play a pivotal role in the onset and progression of hypertension.^3,4 Therefore, simple composite metabolic indices that reflect the overall metabolic burden may help improve early detection and risk stratification for hypertension.

The triglyceride–total cholesterol–body weight index (TCBI) is a newly proposed composite metabolic indicator calculated from three routinely measured parameters: triglycerides (TG), total cholesterol (TC), and body weight (BW). ⁵ In recent years, several studies have suggested that TCBI has potential value in evaluating cardiometabolic health. A prospective study based on the Mashhad Stroke and Heart Atherosclerotic Disorder (MASHAD) cohort reported that higher TCBI was significantly associated with an increased risk of cardiovascular disease (CVD) mortality. In that study, TCBI showed better predictive performance than the traditional atherogenic index of plasma, suggesting that this index may capture a broader lipid- and weight-related cardiometabolic risk profile. ⁶ In addition, recent cohort evidence showed that cumulative TCBI was positively associated with stroke risk, and hypertension mediated approximately 27% of this association. ⁷ This finding indicates that long-term exposure to elevated TCBI may be linked to cerebrovascular events partly through blood pressure-related pathways, thereby providing epidemiological support for a potential connection between TCBI and hypertension.

However, most previous studies have focused on cardiovascular or cerebrovascular outcomes rather than hypertension itself. Moreover, many studies have relied on single-time-point TCBI measurements, which may not fully capture long-term metabolic exposure.^8–10 Given this background, no previous study has directly examined whether TCBI is independently associated with the risk of hypertension. Because TCBI integrates routinely available lipid and BW measures, it may provide a simple indicator of lipid- and weight-related metabolic burden for hypertension risk prediction. Using data from the English Longitudinal Study of Ageing (ELSA), this study evaluated the association between TCBI and hypertension, as well as its predictive value, through cross-sectional, prospective, and trajectory analyses. The findings aimed to clarify whether TCBI could serve as a simple and reproducible metabolic indicator for early hypertension risk stratification and prevention strategies.

Methods

Study design

This study utilized data from the ELSA, an ongoing nationally representative cohort designed to explore the health, social, and psychological aspects of ageing among adults in England. The present study was a secondary analysis of a prospective longitudinal cohort, incorporating cross-sectional, prospective, and trajectory analyses. The reporting of this observational study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. ¹¹ ELSA was initiated in 2002, with participants reinterviewed every 2 years and clinical assessments performed by nurses every 4 years to obtain anthropometric and biochemical data. Because biochemical measures were first available in Wave 2 (2004–2005) that wave served as the baseline for this analysis. For the prospective analysis, participants without hypertension at baseline were followed from Wave 2 (2004–2005) through subsequent ELSA waves for incident hypertension, with a median follow-up duration of 12 years. The present study used publicly available, de-identified ELSA data and was conducted in accordance with the Declaration of Helsinki (1975), as revised in 2024. Ethical approval for the ELSA study was granted by the London Multicentre Research Ethics Committee (MREC/01/2/91), and all participants provided written informed consent. ¹² Because the present analysis was based on publicly available, de-identified data and involved no direct contact with participants, additional local institutional review board approval was not required.

As shown in Figure 1, Wave 2 initially included 9,432 participants. Those with missing hypertension data (n = 2) or information necessary to compute TCBI (n = 3,699) were excluded, leaving 5,731 participants for the cross-sectional analysis. For the longitudinal cohort, individuals with hypertension at baseline (n = 3,174) and those lacking follow-up blood pressure records (n = 205) were excluded, resulting in 2,352 participants eligible for the prospective analysis. For the trajectory modeling, participants missing at least two TCBI measurements or hypertension outcome data (n = 3,009) were further excluded, leaving 2,722 participants included in the final trajectory analysis.

Figure 1.

Flow chart for inclusion and exclusion of the study population.

To assess potential selection bias related to missing data, baseline characteristics were compared between included and excluded participants based on the available data for each variable, and the results are presented in Table S1.

Calculation of TCBI

Blood specimens were obtained by trained nurses using standardized venipuncture techniques and analyzed in the central laboratory of the Royal Victoria Infirmary, Newcastle, UK. ¹³ Serum triglycerides (TG, mg/dL) and (TC, mg/dL) were determined by enzymatic assays, and BW (kg) was measured during the nurse assessment. The triglyceride–total cholesterol–body weight index (TCBI) was calculated as the product of TG, TC, and BW divided by 1,000. All calculated TCBI values were log-transformed before statistical analysis. Therefore, each one-unit increase in log-transformed TCBI represents a relative increase on the original TCBI scale rather than an absolute one-unit increase in the raw TCBI value.

Definition of hypertension

Blood pressure was measured by trained nurses under standardized conditions. Multiple blood pressure readings were obtained during the nurse assessment, and the mean of the valid repeated readings was used for analysis. Hypertension was defined as a mean systolic blood pressure ≥140 mmHg, a mean diastolic blood pressure ≥90 mmHg, self-reported use of antihypertensive medication, or physician-diagnosed hypertension.^14,15 For the prospective analysis, new-onset hypertension was identified among participants without hypertension at baseline when any of these criteria were first met during the follow-up period, with a median follow-up duration of 12 years.

Measurement of covariates

Baseline variables were obtained from standardized questionnaires, structured interviews, and nurse examinations. These questionnaires and interviews were developed and validated as part of ELSA and are publicly available through the ELSA documentation portal. Demographic characteristics included age, sex, marital status (married, never married, or other), and educational attainment (high school or below vs. college or above). Lifestyle variables included smoking (yes/no) and alcohol consumption (yes/no). ¹⁶ Diabetes was identified through self-reported physician diagnosis or by laboratory criteria of fasting plasma glucose ≥7.0 mmol/L or HbA1c ≥ 6.5%. ¹⁷ To further evaluate the independence of the included covariates, potential multicollinearity among TCBI and covariates was assessed using variance inflation factors. All generalized variance inflation factor (GVIF)-adjusted values were below 2, indicating no evident multicollinearity among the included variables (Table S2).

Statistical analysis

Continuous variables were assessed for distributional normality before statistical testing. Normally distributed variables were summarized as mean ± standard deviation and compared using unpaired t tests, whereas non-normally distributed variables were summarized as median with interquartile range (IQR) and compared using Mann–Whitney U tests. Categorical variables were summarized as frequency (percentage) and compared using chi-square tests.

For the cross-sectional analyses, logistic regression models were applied to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for the association between TCBI and prevalent hypertension: Model 1 was unadjusted. Model 2 adjusted for age, sex, education, and marital status. Model 3 further adjusted for smoking, alcohol consumption, and diabetes. For the longitudinal analysis, Cox proportional hazards models with identical covariate structures were used to assess the relationship between baseline TCBI and incident hypertension. Group-based trajectory modeling (GBTM) was used to identify distinct longitudinal patterns of TCBI. Models with different numbers of trajectory groups were fitted and compared, and the final model was selected based on model fit, classification accuracy, group size distribution, and clinical interpretability. Each participant was assigned to the trajectory group with the highest posterior membership probability. Model selection criteria for the trajectory models are presented in Table S3. Restricted cubic spline (RCS) analysis was used to evaluate potential dose–response relationships.

To evaluate the incremental predictive value of TCBI, two prediction models were compared: a covariate model including age, sex, education, marital status, smoking, alcohol consumption, and diabetes, and an expanded model that additionally included TCBI. Model discrimination was assessed using the C-index with 95% CIs, and the statistical significance of the difference between the two models was evaluated using DeLong's test.

Participants with missing exposure or outcome data were excluded from the corresponding analyses. Missing covariate data were addressed using multiple imputation.

All analyses were performed using R (version 4.4.0), and a two-sided p < 0.05 was considered statistically significant.

Results

Cross-sectional analysis

Baseline characteristics

A total of 5,731 participants were included in the cross-sectional analysis, consisting of 3,174 individuals with hypertension and 2,557 without hypertension. Baseline characteristics according to hypertension status are summarized in Table S4. Participants with hypertension were generally older, had higher body mass index (BMI) values, and had a lower proportion of married and highly educated individuals. The prevalence of diabetes was also higher in the hypertensive group. In terms of lifestyle factors, smoking and alcohol consumption were less frequent among participants with hypertension. Notably, the median TCBI level was significantly higher in participants with hypertension compared with the control group (7.78 (7.34–8.24) vs. 7.66 (7.21–8.12), p < 0.001), indicating that elevated TCBI may be associated with a higher prevalence of hypertension.

Association between TCBI and hypertension

As presented in Table 1, TCBI was positively associated with the prevalence of hypertension. In the fully adjusted model, each one-unit increase in TCBI was related to 43% higher odds of hypertension (OR = 1.43, 95% CI: 1.31–1.55). When participants were categorized into quartiles, the odds of hypertension increased steadily across higher TCBI levels. Compared with the lowest quartile, the fully adjusted ORs were 1.31 (95% CI: 1.12–1.52), 1.43 (95% CI: 1.23–1.67), and 1.83 (95% CI: 1.57–2.14) for Q2, Q3, and Q4, respectively, showing a significant linear trend (p for trend < 0.001). These findings suggest that higher TCBI levels were positively associated with the prevalence of hypertension in the cross-sectional analysis.

Table 1.

Association between TCBI and hypertension in the cross-sectional study.

Characteristic	Model 1 OR (95% CI) p value	Model 2 OR (95% CI) p value	Model 3 OR (95% CI) p value
TCBI (continuous)	1.33 (1.23, 1.44) <0.001	1.45 (1.34, 1.58) <0.001	1.43 (1.31, 1.55) <0.001
TCBI quartile
Q1 (5.62–7.28)	Ref	Ref	Ref
Q2 (7.28–7.73)	1.28 (1.14, 1.49) 0.001	1.31(1.13, 1.53) <0.001	1.31 (1.12, 1.52) 0.001
Q3 (7.73–8.19)	1.37 (1.18,1.58) <0.001	1.44 (1.24, 1.68) <0.001	1.43 (1.23, 1.67) <0.001
Q4 (8.19–10.92)	1.64 (1.41,1.90) <0.001	1.90 (1.62, 2.21) <0.001	1.83 (1.57, 2.14) <0.001
p for trend	<0.001	<0.001	<0.001

CI: confidence interval; OR: odds ratio; TCBI: triglyceride–total cholesterol–body weight index.

Model 1: Nonadjusted.

Model 2: Adjusted for age, sex, education, and marital status.

Model 3: Further adjusted for smoke, drinking, and diabetes based on Model 2.

Dose-response relationship between TCBI and hypertension

The RCS curve (Figure S1) illustrated a steady positive association between TCBI and the prevalence of hypertension. The overall relationship was statistically significant (p for overall < 0.001), with no evidence of nonlinearity (p for nonlinearity = 0.723). As TCBI increased, the odds of hypertension gradually increased, indicating a linear dose–response association across the observed range.

Prospective analysis

Baseline characteristics and incident hypertension

After excluding participants with hypertension at baseline and those without follow-up data, a total of 2,352 participants were included in the prospective cohort analysis, comprising 941 individuals who developed hypertension during follow-up and 1,411 who remained normotensive. Baseline characteristics according to hypertension status are summarized in Table S5. Compared with participants who remained normotensive, those who developed hypertension had slightly higher BMI values and a higher prevalence of diabetes. Differences in age, sex, marital status, education, smoking, and alcohol consumption were not statistically significant between the two groups. Importantly, baseline TCBI levels were higher among participants who subsequently developed hypertension than among those who did not (7.69 (7.26–8.17) vs. 7.63 (7.17–8.09), p = 0.018), suggesting that elevated baseline TCBI may predict future hypertension risk.

Association between TCBI and incident hypertension

As shown in Table 2, baseline TCBI was positively associated with the risk of developing hypertension. In the fully adjusted model, each one-unit increase in TCBI was linked to a 16% higher risk of incident hypertension (hazard ratio (HR) = 1.16, 95% CI: 1.05–1.28). When TCBI was divided into quartiles, the risk of hypertension increased progressively across categories, with fully adjusted HRs of 1.33 (95% CI: 1.11–1.60), 1.16 (95% CI: 0.96–1.41), and 1.41 (95% CI: 1.17–1.70) for Q2, Q3, and Q4, respectively, compared with the lowest quartile. A significant linear trend was observed across quartiles (p for trend = 0.003), indicating that higher baseline TCBI levels were associated with a greater likelihood of developing hypertension over time.

Table 2.

Association between TCBI and hypertension in the prospective cohort study.

Characteristic	Model 1 HR (95% CI) p value	Model 2 HR (95% CI) p value	Model 3 HR (95% CI) p value
TCBI (continuous)	1.15 (1.05, 1.27) 0.004	1.17 (1.06, 1.29) 0.001	1.16 (1.05, 1.28) 0.002
TCBI quartile
Q1 (5.88–7.21)	Ref	Ref	Ref
Q2 (7.21–7.65)	1.32 (1.10, 1.59) 0.003	1.33(1.10, 1.60) 0.003	1.33 (1.11, 1.60) 0.002
Q3 (7.65–8.12)	1.17 (0.97,1.42) 0.096	1.16 (0.96, 1.41) 0.115	1.16 (0.96, 1.41) 0.115
Q4 (8.12–10.51)	1.39 (1.16,1.67) <0.001	1.43 (1.19, 1.73) <0.001	1.41 (1.17, 1.70) <0.001
p for trend	0.003	0.002	0.003

CI: confidence interval; HR: hazards ratio; TCBI: triglyceride–total cholesterol–body weight index.

Model 1: Nonadjusted.

Model 2: Adjusted for age, sex, education, and marital status.

Model 3: Further adjusted for smoke, drinking, and diabetes based on model 2.

Dose–response relationship between TCBI and incident hypertension

The RCS analysis (Figure S2) demonstrated a positive and approximately linear association between baseline TCBI and the risk of incident hypertension. The overall relationship was statistically significant (p for overall = 0.010), with no evidence of nonlinearity (p for nonlinearity = 0.752). The risk of hypertension increased gradually as TCBI levels increased, indicating a steady dose–response relationship across the observed range.

Trajectory analysis

TCBI trajectories and baseline characteristics

As shown in Figure 2, GBTM identified three distinct longitudinal TCBI patterns: a low-stable group (n = 544), a moderate-stable group (n = 1,022), and a high-decreasing group (n = 1,156). The trajectories indicated that TCBI levels remained relatively stable within each group, with a slight downward trend observed in the high-decreasing group. Baseline characteristics across trajectory groups are summarized in Table S6. Participants in the higher TCBI trajectory groups were generally older, had higher BMI values, and included a greater proportion of males. The prevalence of smoking and diabetes increased progressively from the low to the high group, whereas marital status, education, and alcohol consumption did not differ significantly among the groups. The incidence of hypertension varied significantly across the trajectory groups, with rates of 50.55% in the low group, 57.05% in the moderate group, and 62.02% in the high group (p < 0.001).

Figure 2.

Longitudinal TCBI trajectories identified using group-based trajectory modeling. The x-axis represents ELSA survey waves, with Wave 2 serving as the baseline. Waves 2, 4, 6, and 8 correspond to 2004–2005, 2008–2009, 2012–2013, and 2016–2017, respectively.

Association between TCBI trajectories and hypertension

As shown in Table 3, higher TCBI trajectory groups were associated with an increased risk of hypertension. In the fully adjusted model, compared with the low-trajectory group, the moderate-trajectory group had 38% higher odds of hypertension (OR = 1.38, 95% CI: 1.11–1.70), and the high-decreasing trajectory group had 69% higher odds of hypertension (OR = 1.69, 95% CI: 1.37–2.09); both associations were statistically significant (p < 0.001). These results indicate a graded association between long-term TCBI trajectories and the risk of hypertension.

Table 3.

Association between TCBI and hypertension in the trajectory analysis cohort.

Characteristic	Model 1 OR (95% CI) p value	Model 2 OR (95% CI) p value	Model 3 OR (95% CI) p value
Low group	Ref	Ref	Ref
Moderate group	1.31 (1.06, 1.62) 0.011	1.36 (1.10, 1.68) 0.004	1.38 (1.11, 1.70) <0.001
High group	1.60 (1.31, 1.97) <0.001	1.72 (1.39, 2.12) <0.001	1.69 (1.37, 2.09) <0.001

CI: confidence interval; OR: odds ratio.

Model 1: Nonadjusted.

Model 2: Adjusted for age, sex, education, and marital status.

Model 3: Further adjusted for smoke, drinking, and diabetes based on model 2.

Sensitivity analysis

To assess the robustness of the findings, two sensitivity analyses were conducted for the cross-sectional analysis. First, subgroup analyses were performed according to major characteristics, including sex, age, smoking, alcohol consumption, diabetes, and BMI. The positive association between TCBI and hypertension remained consistent across all subgroups, with no significant interactions observed (Table S7). Second, propensity score matching (PSM) was performed using the covariates as matching variables to create a 1:1 matched cohort and balance baseline characteristics. After matching, multivariable logistic regression was repeated, and the results remained consistent with the primary analysis, demonstrating a persistent positive association between TCBI and hypertension (Tables S8 and S9). These findings support the robustness of the study conclusions.

Incremental predictive performance of TCBI

To evaluate the incremental predictive value of TCBI beyond conventional covariates, two models were compared: a covariate model including age, sex, education, marital status, smoking, alcohol consumption, and diabetes, and an expanded model that additionally included TCBI. The covariate model without TCBI yielded a C-index of 0.639 (95% CI: 0.625–0.653), whereas the addition of TCBI increased the C-index to 0.654 (95% CI: 0.640–0.668). This improvement in the C-index was statistically significant (p < 0.001).

Discussion

This study is the first to systematically examine the association between the triglyceride–total cholesterol–body weight index (TCBI) and the risk of hypertension among middle-aged and older adults in the United Kingdom using nationally representative data from the ELSA. The results revealed a significant positive association between TCBI and the risk of hypertension, which remained consistent across the cross-sectional, prospective, and trajectory analyses. These findings provide preliminary epidemiological evidence supporting the relevance of TCBI as a metabolic indicator for hypertension risk assessment.

TCBI is a recently proposed composite metabolic index calculated from serum triglycerides, TC, and BW. As an integrated indicator reflecting both lipid metabolism and metabolic reserve, TCBI has potential clinical value in assessing metabolic-related diseases. ¹⁸ Compared with traditional nutritional or metabolic indicators, such as the Geriatric Nutritional Risk Index, Controlling Nutritional Status (CONUT), and Prognostic Nutritional Index (PNI), TCBI relies solely on routine clinical parameters, supporting its accessibility and practical application.^19,20 Recent studies have demonstrated that TCBI is closely associated with various cardiovascular outcomes, including coronary heart disease and stroke,^8,21 and has shown good prognostic value in conditions such as heart failure, coronary artery disease, transcatheter aortic valve replacement, and acute myocardial infarction.^10,22–24 Moreover, TCBI has also been linked to cognitive decline and sarcopenia, indicating that it reflects systemic metabolic homeostasis.^25,26 In general, higher TCBI levels indicate a greater metabolic burden, and their relationship with elevated blood pressure warrants further investigation.

Currently, evidence regarding the relationship between TCBI and hypertension remains limited. Most existing studies have focused on disease prognosis among patients with hypertension, reporting that lower TCBI levels are associated with adverse outcomes such as stroke and renal events.^21,27 In contrast, this study is the first to evaluate TCBI as a predictor of hypertension risk in the general population, highlighting its potential role in primary prevention. Notably, our findings demonstrated a stable linear positive association between TCBI and the risk of hypertension, which was consistent across subgroups stratified by sex, age, smoking, alcohol consumption, and diabetes status (Table S7). This suggests that the association between TCBI and hypertension risk is robust and broadly consistent across these population subgroups. Furthermore, the trajectory analysis indicated that individuals in the high-decreasing TCBI trajectory group had the greatest risk of developing hypertension, supporting the role of cumulative metabolic burden in blood pressure elevation. The downward trend in this group may reflect reductions in triglycerides, TC, or BW during follow-up, potentially related to lifestyle changes, medication use, or selective survival; however, the present data did not allow these explanations to be directly evaluated.

The mechanisms linking TCBI and hypertension may involve several metabolic and vascular pathways. High TCBI reflects the combined burden of elevated triglycerides, elevated TC, and greater BW. Dyslipidemia may impair endothelial function and promote oxidative stress, arterial stiffness, and vascular remodeling, all of which can increase peripheral vascular resistance and contribute to higher blood pressure.^28,29 Excess BW may further increase blood pressure through obesity-related inflammation, insulin resistance, sympathetic nervous system activation, and stimulation of the renin–angiotensin–aldosterone system (RAAS).^30,31 These processes may reinforce one another. Lipid abnormalities and adiposity can jointly promote chronic low-grade inflammation, endothelial injury, and impaired vascular relaxation.^32,33 Therefore, TCBI may serve as a simple marker that captures the combined effects of lipid metabolism and obesity-related metabolic stress, which may partly explain its association with the risk of hypertension.

The present findings have important clinical implications. Although the between-group difference in median TCBI appeared modest, TCBI was log-transformed before analysis; therefore, this difference should be interpreted on the logarithmic scale, and the clinical relevance of TCBI is better reflected by its consistent associations across the multivariable, quartile-based, prospective, and trajectory analyses. First, the observed linear dose–response relationship supports the potential use of TCBI as a simple, quantifiable index for hypertension risk prediction and population risk stratification. The incremental discrimination analysis showed that adding TCBI to the covariate model increased the C-index from 0.639 to 0.654, with a statistically significant difference between the two models. This suggests that TCBI may provide modest additional discriminatory information beyond conventional risk factors, although its clinical utility requires further validation. Second, the consistency of the results across multiple analytical levels—including cross-sectional, longitudinal, and trajectory analyses—along with the sensitivity analyses (PSM and subgroup analyses), strengthens the robustness of our conclusions and suggests that TCBI reflects metabolic burden associated with blood pressure regulation over time. Finally, when considered together with previous evidence linking TCBI to cardiovascular and cerebrovascular outcomes, our findings further underscore the potential value of TCBI in the early identification and prevention of hypertension within the broader context of metabolic syndrome management.

This study has some limitations. First, as an observational analysis, causal relationships cannot be fully established; future studies incorporating randomized controlled trials or Mendelian randomization are needed to confirm causality. Second, potential confounders such as dietary intake and physical activity were not included, which may have introduced residual confounding. Smoking and alcohol consumption were treated as binary variables, which may not fully capture exposure intensity, duration, frequency, or cumulative lifestyle burden. Although diabetes was adjusted for in the multivariable models and no evident multicollinearity was observed in the variance inflation factor analysis, residual confounding related to diabetes duration, severity, treatment, or glycemic control cannot be completely excluded. Third, although baseline characteristics were compared between included and excluded participants using the available data, differences in several variables were observed, suggesting that potential selection bias due to missing data cannot be completely excluded. Fourth, because hypertension status was assessed at discrete follow-up visits, the exact timing of hypertension onset could not be determined, resulting in potential interval censoring in the prospective analysis. Finally, because the study population consisted of middle-aged and older adults in the United Kingdom, the generalizability of these findings to other ethnic groups or younger populations requires further investigation.

Conclusion

This study demonstrated a consistent positive association between TCBI levels and the risk of hypertension among middle-aged and older adults in the United Kingdom. The relationship remained robust across the cross-sectional, prospective, and trajectory analyses. As a simple and reproducible metabolic indicator derived from routine clinical measurements, TCBI shows promise for the early identification and risk stratification of hypertension, providing new evidence to support precision prevention and management strategies.