ABSTRACT
We aimed to examine the individual and combined impact of prehypertension, prediabetes, and predyslipidemia on all‐cause and cardiovascular disease (CVD) mortality in community‐dwelling adults. A retrospective cohort study of 11 986 US adults from the 1999–2018 National Health and Nutrition Examination Survey was conducted. Participants were categorized into four mutually exclusive groups based on the cumulative number of these conditions. Multivariate Cox proportional hazards models were applied to calculate the hazard ratios (HRs) and 95% confidence intervals (CIs) for mortality outcomes. The prevalences of no, one, two, and all three preclinical conditions were 37.20%, 38.39%, 19.86%, and 4.55%, respectively. Over a median follow‐up of 10 years, 636 (3.83%) deaths occurred, including 170 (26.73%) from CVD. In a dose‐response manner, the adjusted HRs (95%CIs) for all‐cause mortality among those with one, two, and three conditions were 1.87 (1.38–2.52), 2.49 (1.80–3.44), and 2.62 (1.76–3.90), respectively, compared to those with none. The corresponding HRs (95%CIs) for CVD mortality were 3.09 (1.66–5.78), 3.89 (2.06–7.36), and 4.20 (2.04–8.69), respectively. Thus, an increasing cumulative number of preclinical conditions is associated with graded elevated risk of all‐cause and CVD mortality, underscoring the potential for early intervention during the preclinical phase to improve long‐term health outcomes.
Keywords: cohort study, mortality, NHANES, prediabetes, predyslipidemia, prehypertension
1. Introduction
Hypertension, diabetes, and dyslipidemia are well‐established major metabolic risk factors for cardiovascular disease (CVD) and its associated mortality, contributing to a substantial global healthcare burden [1]. These conditions typically evolve progressively through identifiable precursor stages, including prehypertension, prediabetes, and predyslipidemia. Studies have demonstrated that up to two‐thirds of individuals with prehypertension progress to hypertension within 4 years [2], and approximately 12.5% of individuals with prediabetes progress to diabetes within 10 years [3]. Importantly, these preclinical entities are potentially reversible with appropriate lifestyle or pharmacological interventions [4]. Consequently, the identification and proper management of these preclinical stages hold significant promise for reducing CVD morbidity and mortality.
Studies on the associations between these preclinical conditions and mortality risk have yielded mixed and conflicting results. For example, a meta‐analysis of 20 prospective cohort studies showed that prehypertension significantly increased CVD mortality but not all‐cause mortality [5]. However, a recent large‐scale epidemiological study demonstrated that prehypertension increased the risk of all‐cause and CVD mortality by 27% and 54%, respectively, in individuals with diabetes but not in those without diabetes [6]. Similarly, although a meta‐analysis of 53 prospective cohort studies with 1 611 339 individuals found that prediabetes increased all‐cause mortality by 13% [7], a study by Tang et al. found no significant association between prediabetes and all‐cause and CVD mortality in older adults [8]. Data on the relationship between predyslipidemia and mortality risk are scarce. Given that epidemiological studies have indicated a high prevalence of coexisting preclinical conditions even in apparently healthy individuals [9], examining their joint effects on long‐term survival is of critical clinical importance.
To the best of our knowledge, no studies have investigated the associations between cumulative burden of preclinical conditions and mortality risk. To address this knowledge gap, the current study aimed to examine the hypothesis that an increasing cumulative number of preclinical conditions would be associated with a gradual increase in the risk of all‐cause and CVD mortality.
2. Methods
2.1. Participants Selection and Ethics Declaration
Data were sourced from the publicly available National Health and Nutrition Examination Survey, a continuous series of nationwide cross‐sectional surveys conducted every two years to assess the health and nutritional status of the US population. The detailed methodology has been published elsewhere [10, 11]. The NHANES was approved by the Institutional Review Board of the National Center for Health Statistics, and all participants provided informed consent.
This study included adult participants (age ≥ 20 years) from the 1999 to 2018 survey cycles. The exclusion criteria were as follows: (1) current pregnancy or a self‐reported history of cancer; (2) a diagnosis of diabetes, hypertension, or dyslipidemia; (3) missing data on preclinical conditions or mortality follow‐up; and (4) missing data of covariates. As shown in Figure 1, the final analytic sample comprised 11,986 participants.
FIGURE 1.
Flowchart showing the process of participant selection and exclusion from the 1999 to 2018 National Health and Nutrition Examination Survey (NHANES).
2.2. Definitions
Definitions for diabetes, hypertension, and dyslipidemia were consistent with those commonly used in studies of the NHANES population. In brief, diabetes was defined by self‐reported diagnosis, or a fasting blood glucose ≥ 7.0 mmol/L, or an HbA1c ≥ 6.5%, or a 2‐hour post‐load glucose level ≥ 11.1 mmol/L, or current use of insulin or oral hypoglycemic medications [11]. Hypertension was defined as self‐report, or use of antihypertensive medications, or an averaged measured blood pressure at the Mobile Examination Center ≥ 140/90 mmHg. Dyslipidemia was defined by self‐report, current use of lipid‐lowering medications, or a triglyceride ≥ 200 mg/dL, or a total cholesterol ≥ 240 mg/dL, or a low‐density lipoprotein cholesterol ≥ 160 mg/dL, or a high‐density lipoprotein cholesterol < 40 mg/dL in men or < 50 mg/dL in women [12].
Accordingly, the diagnostic criteria for prediabetes were based on the criteria established by the American Diabetes Association: a fasting blood glucose of 5.6–6.9 mmol/L, an HbA1c of 5.7%–6.4%, or a 2‐hour post‐load oral glucose tolerance test blood glucose level of 7.8–11.0 mmol/L [13]. Prehypertension was defined as a systolic blood pressure of 120–139 mmHg or a diastolic blood pressure of 80–89 mmHg [14]. Criteria for predyslipidemia included a low‐density lipoprotein cholesterol level of 130–159 mg/dL, or a total cholesterol level of 200–239 mg/dL, or a triglyceride level of 150–199 mg/dL [15]. Participants with isolated low high‐density lipoprotein cholesterol without concurrent elevation of any of the above parameters were not classified as predyslipidemia in this study.
2.3. Outcomes
We focused on all‐cause and CVD mortality in the current study. Survival status and cause of death information were obtained from the National Death Index public‐use files, with follow‐up data available through December 31, 2019. The follow‐up duration for each participant was calculated as the period between Mobile Examination Center examination date and December 31, 2019 or death, whichever occurred first. CVD mortality was identified by International Statistical Classification of Diseases and Related Health Problems 10th Revision codes I00‐I09, I11, I13, I20‐I51, and I60‐I69 [16].
2.4. Covariates
We controlled for the following potential confounding variables in the analyses: sociodemographic characteristics (age, sex, race, marital status, educational attainment, and family income), lifestyle factors (body mass index, physical activity, smoking, and drinking), comorbidity of CVD, kidney function, and use of anti‐platelet medications. Race was self‐reported and categorized into non‐Hispanic White, non‐Hispanic Black, Mexican Americans, other Hispanics, and other races. Marital status was dichotomized into single (divorced, separated, widowed, or never married) and non‐single (married or living with a partner). Family income was assessed by the poverty‐income ratio. Physical activity was categorized as no, moderate, or vigorous. A history of CVD was defined by a self‐reported diagnosis of angina, myocardial infarction, congestive heart failure, coronary heart disease, and stroke. Kidney function was assessed by the estimated glomerular filtration rate, which was calculated using the 2009 Chronic Kidney Disease Epidemiology Collaboration equation based on serum creatinine [17].
2.5. Statistical Analysis
Participants were categorized into four groups based on the cumulative number of prediabetes, prehypertension, and predyslipidemia present, using those with none as the reference group. Baseline characteristics of the study population were summarized using descriptive statistics. The association between the cumulative number of preclinical conditions and mortality risk was assessed using Cox proportional hazards regression models, accounting for the complex survey design of NHANES. The proportional hazards assumption was tested using Schoenfeld residuals. Three sequential models were generated. Model 1 was unadjusted. Model 2 was adjusted for sociodemographic variables, including age, sex, race, marital status, educational attainment, and family income. Model 3 was the fully adjusted model, which included all covariates from Model 2, plus body mass index, physical activity, smoking, drinking, history of CVD, estimated glomerular filtration rate, and use of antiplatelet medications. Results are presented as hazard ratios (HRs) with their corresponding 95% confidence intervals (CIs). A test for trend was performed by treating the ordinal exposure variable as a continuous term in the fully adjusted model. Subgroup analysis, stratified by age, sex, race, body mass index, and physical activity, was conducted. Interaction terms between the cumulative number of preclinical conditions and each subgroup variable were tested for significance. To evaluate the robustness of the findings, sensitivity analysis was performed by excluding participants who died within the first 24 months of follow‐up to mitigate potential reverse causality. A two‐tailed p value of less than 0.05 was considered statistically significant for all analyses.
3. Results
3.1. Comparison of Baseline Characteristics
After applying the inclusion and exclusion criteria, 11 986 participants were included in the analysis. This cohort comprised 37.20% (n = 4304) with no preclinical conditions, 38.39% (n = 4571) with one, 19.86% (n = 2479) with two, and 4.55% (n = 632) with all three preclinical conditions. As presented in Table 1, participants with all three preclinical conditions were significantly older, more likely to be single men, less educated, and physically inactive compared to those with no pre‐disease. They also had a higher prevalence of former smoking and former drinking, CVD, and anti‐platelet medication use.
TABLE 1.
Comparison of baseline characteristics among US adults with different cumulative number of prehypertension, prediabetes, and predyslipidemia.
| Total (n = 11986) | 0 (n = 4304) | 1 (n = 4571) | 2 (n = 2479) | 3 (n = 632) | p value | |
|---|---|---|---|---|---|---|
| Age, years | 39.26 ± 0.22 | 33.63 ± 0.25 | 39.85 ± 0.30 | 46.01 ± 0.44 | 50.98 ± 0.65 | < 0.001 |
| Sex, n (%) | < 0.001 | |||||
| Men | 5936 (49.10) | 1668 (38.48) | 2386 (52.80) | 1492 (59.27) | 390 (60.33) | |
| Women | 6050 (50.90) | 2636 (61.52) | 2185 (47.20) | 987 (40.73) | 242 (39.67) | |
| Race, n (%) | 0.33 | |||||
| Non‐Hispanic White | 5405 (69.15) | 1974 (68.62) | 2078 (69.37) | 1091 (70.11) | 262 (67.50) | |
| Non‐Hispanic Black | 2345 (10.15) | 786 (9.77) | 921 (10.28) | 506 (10.39) | 132 (11.13) | |
| Mexican American | 2035 (8.38) | 742 (8.85) | 739 (7.99) | 437 (8.11) | 117 (8.91) | |
| Other Hispanic | 959 (5.54) | 340 (6.16) | 349 (5.17) | 214 (5.30) | 56 (4.71) | |
| Other races | 1242 (6.78) | 462 (6.60) | 484 (7.19) | 231 (6.09) | 65 (7.75) | |
| Marital status, n (%) | < 0.001 | |||||
| Non‐single | 7115 (62.45) | 2396 (58.53) | 2704 (62.93) | 1580 (66.81) | 435 (71.56) | |
| Single | 4871 (37.55) | 1908 (41.47) | 1867 (37.07) | 899 (33.19) | 197 (28.44) | |
| Poverty‐income ratio | 3.17±0.03 | 3.05±0.04 | 3.20±0.04 | 3.31±0.05 | 3.27±0.09 | < 0.001 |
| Education, n (%) | < 0.001 | |||||
| < High school | 834 (3.53) | 251 (3.25) | 282 (3.19) | 216 (4.06) | 85 (6.43) | |
| High school | 4001 (29.52) | 1306 (26.63) | 1593 (30.50) | 870 (32.01) | 232 (34.03) | |
| > High school | 7151 (66.95) | 2747 (70.12) | 2696 (66.31) | 1393 (63.93) | 315 (59.54) | |
| Physical activity, n (%) | 0.006 | |||||
| No | 5613 (42.32) | 2023 (43.44) | 2085 (40.77) | 1164 (41.08) | 341 (51.53) | |
| Moderate | 2881 (25.49) | 994 (24.52) | 1141 (25.93) | 611 (27.24) | 135 (22.02) | |
| Vigorous | 3492 (32.19) | 1287 (32.04) | 1345 (33.30) | 704 (31.68) | 156 (26.45) | |
| Body mass index, kg/m2 | 26.25±0.08 | 24.73±0.09 | 26.56±0.12 | 27.94±0.15 | 28.68±0.25 | < 0.001 |
| Smoking, n (%) | < 0.001 | |||||
| Never | 7232 (59.55) | 2786 (63.51) | 2739 (58.90) | 1366 (55.01) | 341 (52.61) | |
| Former | 2285 (20.48) | 627 (16.39) | 886 (20.90) | 615 (25.99) | 157 (26.31) | |
| Current | 2469 (19.97) | 891 (20.10) | 946 (20.20) | 498 (19.00) | 134 (21.08) | |
| Drinking, n (%) | < 0.001 | |||||
| Never | 1450 (9.57) | 550 (10.32) | 542 (9.18) | 287 (9.30) | 71 (7.89) | |
| Former | 1421 (10.22) | 414 (8.75) | 540 (9.76) | 338 (11.75) | 129 (19.43) | |
| Mild | 3976 (34.94) | 1312 (31.68) | 1564 (36.43) | 896 (38.25) | 204 (34.56) | |
| Moderate | 2194 (20.08) | 884 (22.38) | 809 (19.40) | 416 (18.12) | 85 (15.60) | |
| Heavy | 2945 (25.19) | 1144 (26.87) | 1116 (25.23) | 542 (22.58) | 143 (22.52) | |
| CVD, n (%) | 336 (2.41) | 72 (1.53) | 137 (2.73) | 104 (3.16) | 23 (3.53) | < 0.001 |
| eGFR, ml/min/1.73m2 | 101.31±0.34 | 106.07±0.44 | 100.59±0.44 | 95.74±0.55 | 92.77±0.92 | < 0.001 |
| Anti‐platelet use, n (%) | 69 (0.46) | 10 (0.18) | 30 (0.54) | 25 (0.85) | 4 (0.42) | 0.01 |
| Prediabetes, n (%) | 2952 (21.73) | 0 (0) | 952 (18.60) | 1368 (50.54) | 632 (100.00) | < 0.001 |
| Prehypertension, n (%) | 4155 (34.08) | 0 (0) | 1706 (38.82) | 1817 (73.62) | 632 (100.00) | < 0.001 |
| Predyslipidemia, n (%) | 4318 (35.96) | 0 (0) | 1913 (42.58) | 1773 (75.84) | 632 (100.00) | < 0.001 |
Note: The p values denote the comparison among the four groups.
Abbreviations: CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate.
3.2. Survival Outcomes
During a median follow‐up of 10 years (interquartile range: 5.6–14.7), 636 (3.83%) participants died, of which 170 (26.73%) deaths were attributable to CVD. All‐cause mortality rates increased with the number of preclinical conditions: 1.78% for none, 4.10% for one, 6.33% for two, and 7.47% for three. Similarly, corresponding CVD mortality rates were 0.26%, 1.14%, 1.69%, and 2.03%, respectively. Kaplan–Meier survival curves (Figure 2) indicated that participants with all three preclinical conditions had the poorest survival probability, while those with no preclinical conditions had the best.
FIGURE 2.
Kaplan–Meier survival curves by cumulative number of pre‐disease states. Kaplan–Meier curves illustrate the probability of survival free from all‐cause mortality (A) and cardiovascular disease (B) mortality, stratified by the cumulative number of preclinical conditions (0, 1, 2, or 3) among participants from the 1999 to 2018 National Health and Nutrition Examination Survey. The number of participants at risk over time is presented at the risk table below each graph. The difference in survival probability among the four groups was statistically significant (Log‐rank test, p < 0.001).
3.3. Associations Between Cumulative Number of Preclinical Conditions and Mortality
We first explored the association between each preclinical condition with mortality outcomes. As shown in Table 2, the multivariate‐adjusted Cox regression analysis showed that prehypertension and prediabetes were significantly associated with an increased risk of all‐cause and CVD mortality. Predyslipidemia, however, was associated only with CVD mortality (HR = 2.37, 95%CI 1.04–5.40), not with all‐cause mortality (HR = 1.22, 95%CI 0.82–1.83).
TABLE 2.
Associations between individual preclinical condition with all‐cause and cardiovascular disease mortality among US adults from the 1999 to 2018 National Health and Nutrition Examination Survey.
| Model 1 | Model 2 | Model 3 | ||||
|---|---|---|---|---|---|---|
| HR (95% CI) | p value | HR (95%CI) | p value | HR (95%CI) | p value | |
| All‐cause mortality | ||||||
| Normal | 1 (Reference) | / | 1 (Reference) | / | 1 (Reference) | / |
| Prehypertension | 2.72 (1.97–3.75) | <0.001 | 2.46 (1.74–3.49) | <0.001 | 2.09 (1.50–2.92) | <0.001 |
| Prediabetes | 4.86 (3.39–6.99) | <0.001 | 4.57 (3.14–6.66) | <0.001 | 3.10 (2.15–4.48) | <0.001 |
| Predyslipidemia | 1.33 (0.89–1.99) | 0.16 | 1.38 (0.93–2.07) | 0.11 | 1.22 (0.82–1.83) | 0.33 |
| Cardiovascular disease mortality | ||||||
| Normal | 1 (Reference) | / | 1 (Reference) | / | 1 (Reference) | / |
| Prehypertension | 4.29 (2.11–8.69) | <0.001 | 3.70 (1.79–7.65) | <0.001 | 2.85 (1.43–5.69) | 0.003 |
| Prediabetes | 10.94 (5.35–22.37) | <0.001 | 9.96 (4.77–20.76) | <0.001 | 5.64 (2.63–12.08) | <0.001 |
| Predyslipidemia | 2.75 (1.21–6.25) | 0.02 | 2.84 (1.27–6.37) | 0.01 | 2.37 (1.04–5.40) | 0.04 |
Note: Model 1 was unadjusted. Model 2 was adjusted for age, sex, race, poverty‐income ratio, marital status, and education level. Model 3 was further adjusted for body mass index, physical activity, smoking, drinking, cardiovascular disease, anti‐platelet medication use, and estimated glomerular filtration rate.
Abbreviations: CI, confidence interval; HR, hazard ratio.
The associations between the cumulative number of pre‐diseases and mortality risk are summarized in Table 3. In the crude model, compared to the no pre‐disease group, participants with one, two and all three preclinical conditions had a 1.33‐, 2.72‐, and 4.00‐fold increased risk of all‐cause mortality, and a 3.37‐, 5.69‐, and 8.15‐fold increased risk of CVD mortality, respectively. These associations remained significant after full adjustment, though the HRs were attenuated. The fully adjusted HRs (95% CIs) for all‐cause mortality were 1.87 (1.38–2.52), 2.49 (1.80–3.44), and 2.62 (1.76–3.90) for one, two, and three preclinical conditions, respectively. The corresponding HRs for CVD mortality were 3.09 (1.66–5.78), 3.89 (2.06–7.36), and 4.20 (2.04–8.69).
TABLE 3.
Associations between cumulative number of pre‐diseases with all‐cause and cardiovascular disease mortality among US adults from the 1999–2018 National Health and Nutrition Examination Survey.
| Model 1 | Model 2 | Model 3 | ||||
|---|---|---|---|---|---|---|
| HR (95% CI) | p value | HR (95%CI) | p value | HR (95%CI) | p value | |
| All‐cause mortality | ||||||
| 0 | 1 (Reference) | / | 1 (Reference) | / | 1 (Reference) | / |
| 1 | 2.33 (1.73–3.13) | <0.001 | 2.23 (1.64–3.04) | <0.001 | 1.87 (1.38–2.52) | <0.001 |
| 2 | 3.72 (2.69–5.14) | <0.001 | 3.49 (2.50–4.88) | <0.001 | 2.49 (1.80–3.44) | <0.001 |
| 3 | 5.00 (3.36–7.43) | <0.001 | 4.30 (2.83–6.54) | <0.001 | 2.62 (1.76–3.90) | <0.001 |
| P for trend | <0.001 | <0.001 | <0.001 | |||
| Cardiovascular disease mortality | ||||||
| 0 | 1 (Reference) | / | 1 (Reference) | / | 1 (Reference) | / |
| 1 | 4.37 (2.35–8.12) | <0.001 | 4.10 (2.19–7.67) | <0.001 | 3.09 (1.66–5.78) | <0.001 |
| 2 | 6.69 (3.60–12.44) | <0.001 | 6.13 (3.28–11.48) | <0.001 | 3.89 (2.06–7.36) | <0.001 |
| 3 | 9.15 (4.52–18.49) | <0.001 | 7.50 (3.71–15.18) | <0.001 | 4.20 (2.04–8.69) | <0.001 |
| P for trend | <0.001 | <0.001 | <0.001 | |||
Note: Model 1 was unadjusted. Model 2 was adjusted for age, sex, race, poverty‐income ratio, marital status, and education level. Model 3 was further adjusted for body mass index, physical activity, smoking, drinking, cardiovascular disease, anti‐platelet medication use, and estimated glomerular filtration rate.
Abbreviations: CI, confidence interval; HR, hazard ratio.
An analysis of different combinations of two preclinical conditions suggested that the combination of prediabetes and prehypertension was associated with the highest excess mortality (Table S1).
3.4. Subgroup Analysis
Subgroup analyses (Figure 3) showed that the relationships between the cumulative number of pre‐diseases and all‐cause and CVD mortality were not significantly modified by age, sex, race, body mass index, or physical activity level (all P for interaction > 0.05).
FIGURE 3.
Subgroup analysis of the association between cumulative number of preclinical conditions and mortality. Forest plot displaying hazard ratios (HRs) and 95% confidence intervals for all‐cause mortality (A) and cardiovascular disease mortality (B), stratified by participant characteristics. Analyses are based on data from the 1999 to 2018 National Health and Nutrition Examination Survey. The size of the data markers corresponds to the precision of the estimate (inverse of the variance). The dashed vertical line represents the reference (HR = 1.0).
3.5. Sensitivity Analysis
Compared to participants with no preclinical condition, the fully adjusted HRs for all‐cause mortality were 2.07 (1.50–2.84) for one preclinical condition, 2.90 (2.04–4.13) for two, and 3.19 (2.07–4.91) for three (P for trend < 0.001, Table S2). Similarly, for CVD mortality, the HRs were 2.81 (1.49–5.30), 3.71 (1.89–7.28), and 4.33 (2.08–9.02), respectively (P for trend < 0.001). These results reinforce a dose‐response relationship, indicating a gradually elevated mortality risk with an increasing number of preclinical conditions.
4. Discussion
This retrospective cohort study analyzed nationally representative data from community‐dwelling adults to investigate the individual and combined effects of prehypertension, prediabetes, and predyslipidemia on long‐term outcomes. The analysis revealed that all three preclinical conditions were associated with a significantly elevated risk of CVD mortality. Furthermore, prehypertension and prediabetes were also independently associated with an increased risk of all‐cause mortality. Subsequent examinations revealed a clear dose‐response relationship, demonstrating a combined effect of these preclinical conditions on mortality. The findings remained robust and consistent across subgroup and sensitivity analyses. These results underscore the critical importance of early clinical interventions targeting preclinical conditions to improve long‐term patient outcomes.
While previous studies have explored the prognostic implications of prehypertension and prediabetes, findings are frequently inconsistent. For example, Mainous et al. found that prehypertension did not increase all‐cause or CVD mortality [18]. Similarly, the Ohsaki Study of 12 928 Japanese adults reported a population attributable fraction for prehypertension on CVD mortality of merely 7% in middle‐aged adults and 0% in the elderly [19]. In contrast, the San Antonio Heart Study reported that prehypertension increased all‐cause and CVD mortality by 49% and 79%, respectively [20]. Other studies documented that prehypertension alone did not significantly increase mortality rates but, when concurrent with diabetes, could increase all‐cause and CVD mortality [21]. Paradoxically, Ren and associates demonstrated in the community‐based Kailuan Study that the HRs (95% CIs) for prehypertension were 1.27 (1.17–1.38) for all‐cause mortality and 1.54 (1.38–1.71) in participants without diabetes, but 0.88 (0.73–1.07) for all‐cause mortality and 1.20 (0.93–1.56) in those with diabetes [6], indicating that diabetes may nullify the association between prehypertension and mortality risk.
The impact of prediabetes on mortality risk is well‐documented. For instance, an umbrella review of 95 meta‐analyses confirmed that prediabetes is associated with a 5%–25% increase in all‐cause mortality risk, depending on the definitions used, in the general population [22]. Moreover, this elevated mortality risk persisted even in individuals who reverted from prediabetes to normoglycemia [23]. In contrast, the association between predyslipidemia and survival remains largely unknown. Our study found that although predyslipidemia significantly increased CVD mortality, it was not associated with elevated all‐cause mortality. This negative finding might be attributed to a limited follow‐up duration, competing risks of non‐CVD mortality, or the effects of treatment and risk modification.
An intriguing observation of this study is the combined effect of cumulative burden of prehypertension, prediabetes, and predyslipidemia on mortality outcomes. This finding demonstrates that the combined impact of these preclinical conditions is greater than the sum of their individual effects. This concept is supported by Huang et al., who reported that the HRs (95% CIs) for all‐cause mortality were 1.04 (0.88–1.24) for prehypertension alone, 0.96 (0.76–1.21) for prediabetes alone, and 1.19 (0.98–1.46) for combined prehypertension and prediabetes; for CVD mortality, the corresponding HRs (95% CIs) were 1.51 (0.83–2.77), 1.40 (0.64–3.06), and 1.70 (0.88–3.27), respectively, for CVD mortality [24]. The amplified risk for CVD mortality is biologically plausible, as these intermediate states collectively promote accelerated atherosclerosis, endothelial dysfunction, and chronic inflammation. Interestingly, an earlier study demonstrated that the combination of prehypertension and prediabetes predicts the incidence and progression of myocardial infarction, even among apparently healthy individuals [25]. Compared to those with normotension and normoglycemia, participants with coexistent prehypertension and prediabetes exhibited the worst cardiometabolic profiles, such as larger waist circumference, higher insulin resistance, higher triglycerides, and lower high‐density lipoprotein cholesterol [26]. Echoing this notion, Ma's group demonstrated that Chinese individuals with one, two, and all three preclinical conditions had a 33%, 37%, and 55% increased risk of incident CVD over a median of 7.1 years [15]. Moreover, the significant effect on all‐cause mortality further underscores their substantial threat to overall health. This result demonstrates that the clinical burden begins well before the onset of overt disease, underscoring the urgent need for integrated, multifactorial lifestyle and public health interventions targeted at this high‐risk population in the pre‐disease state.
Among participants with two preclinical conditions, the combination of prediabetes and prehypertension was associated with the highest risk of CVD mortality. This is likely attributable to the potent interaction between insulin resistance and heightened arterial pressure, which collectively accelerate endothelial dysfunction and atherosclerosis [27]. In contrast, the combination of prehypertension and predyslipidemia was associated with the lowest risk, possibly because the atherogenic effects of dyslipidemia are often more readily managed with lipid‐lowering treatment [28]. This observed risk hierarchy underscores that not all preclinical condition pairs carry equivalent risk; the specific interaction of metabolic and hemodynamic disturbances in prediabetes and prehypertension is particularly deleterious. These findings identify this confluence as a critical target for intensive preventive strategies.
This study has several strengths, including a large sample size, extended follow‐up duration, and standardized data collection protocols. The robustness of the principal findings were further corroborated by the subgroup and sensitivity analysis. Despite these strengths, several limitations should be taken into consideration. First, all data were collected at baseline, precluding dynamic assessment of disease progression over time. Accordingly, we lacked information on the proportion of participants who developed overt hypertension, diabetes, or dyslipidemia during follow‐up. Medical treatment initiated upon progression to overt disease may have subsequently modified disease trajectories and, by extension, mortality risk [29]. Thus, longitudinal monitoring with repeated measurements would be necessary to more precisely characterize the relationship between preclinical conditions and long‐term mortality [30]. Second, the observational design cannot establish a causal relationship between preclinical conditions and mortality. Third, as the analysis relied on baseline data from a cross‐sectional survey, information on specific management protocols that could have influenced long‐term outcomes was unavailable. Fourth, despite efforts to adjust for multiple covariates, residual confounding remains possible and may have led to an overestimation of the observed associations. Potential unmeasured confounders include dietary patterns, sleep quality, psychological stress, family history of CVD, and medication adherence among participants on antiplatelet therapy. Fifth, despite the large overall sample size, certain subgroups in the stratified analysis had limited sample sizes and were underpowered, as indicated by the wide 95% CIs. This was particularly evident for the “Other race” category and the body mass index ≥30 kg/m2 subgroup. These results should therefore be interpreted with caution. Ultimately, the generalizability of these findings from a single‐country analysis to other populations with different metabolic profiles or genetic backgrounds still warrants further investigation.
5. Conclusions
In conclusion, this community‐based study demonstrates that prehypertension and prediabetes are significantly associated with increased all‐cause and CVD mortality, while predyslipidemia is linked to elevated CVD mortality. Furthermore, a clear dose‐response relationship was observed between the cumulative number of these preclinical conditions and mortality risk. These findings underscore the critical importance of early, integrated interventions targeting these metabolic risk factors to reduce public health burden and improve long‐term survival.
Author Contributions
Jiao Meng and Gang Wang contributed to data analysis and writing of the original manuscript. Wei Wang and Junnan Wu contributed to conceptualization and critical revision. All authors have read and approved the manuscript.
Funding
The authors have nothing to report.
Ethics Statement
The NHANES protocol was approved by the National Center for Health Statistics
Consent
All participants provided informed consent.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Supplementary Tables: jch70264‐sup‐0001‐tableS1‐S2.doc
Acknowledgments
The authors have nothing to report.
Contributor Information
Wei Wang, Email: 13211010060@fudan.edu.cn.
Junnan Wu, Email: junnan.wu@zju.edu.cn.
Data Availability Statement
The dataset for this study is publicly available at CDC website at https://wwwn.cdc.gov/nchs/nhanes/Default.aspx
References
- 1. Vaduganathan M., Mensah G. A., Turco J. V., Fuster V., and Roth G. A., “The Global Burden of Cardiovascular Diseases and Risk: A Compass for Future Health,” Journal of the American College of Cardiology 80, no. 25 (2022): 2361–2371, 10.1016/j.jacc.2022.11.005. [DOI] [PubMed] [Google Scholar]
- 2. Julius S., Nesbitt S. D., Egan B. M., et al., “Feasibility of Treating Prehypertension With an Angiotensin‐Receptor Blocker,” New England Journal of Medicine 354, no 16 (2006): 1685–1697, 10.1056/NEJMoa060838. [DOI] [PubMed] [Google Scholar]
- 3. Davoodian N., Lotfaliany M., Huxley R. R., et al., “Prediabetes Transitions to Normoglycaemia or Type 2 Diabetes and Associated Risk Factors in the Obesity, Diabetes and Cardiovascular Disease Collaboration: An Individual‐Level Pooled Analysis of 19 Prospective Cohort Studies,” Lancet Global Health 13, no 9 (2025): e1533–e1542, 10.1016/S2214-109X(25)00237-2. [DOI] [PubMed] [Google Scholar]
- 4. Galaviz K. I., Weber M. B., Suvada K. B. S., et al., “Interventions for Reversing Prediabetes: A Systematic Review and Meta‐Analysis,” American Journal of Preventive Medicine 62, no 4 (2022): 614–625, 10.1016/j.amepre.2021.10.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Huang Y., Su L., Cai X., et al., “Association of All‐Cause and Cardiovascular Mortality With Prehypertension: A Meta‐Analysis,” American Heart Journal 167, no 2 (2014): 160–168.e1, 10.1016/j.ahj.2013.10.023. [DOI] [PubMed] [Google Scholar]
- 6. Ren Y., Zuo Y., Wang A., et al., “Diabetes Modifies the Association of Prehypertension With Cardiovascular Disease and All‐Cause Mortality,” Journal of Clinical Hypertension 23 no 6 (2021): 1221–1228, 10.1111/jch.14246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Huang Y., Cai X., Mai W., Li M., and Hu Y., “Association Between Prediabetes and Risk of Cardiovascular Disease and all Cause Mortality: Systematic Review and Meta‐Analysis,” BMJ: British medical journal / British Medical Association 355 (2016): i5953, 10.1136/bmj.i5953. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Tang O., Matsushita K., Coresh J., et al., “Mortality Implications of Prediabetes and Diabetes in Older Adults,” Diabetes Care 43, no 2 (2020): 382–388, 10.2337/dc19-1221. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Wu J., Yan W. H., Qiu L., et al., “High Prevalence of Coexisting Prehypertension and Prediabetes Among Healthy Adults in Northern and Northeastern China,” BMC Public Health 11, (2011):794, 10.1186/1471-2458-11-794. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Kong X. and Wang W., “Estimated Glucose Disposal Rate and Risk of Cardiovascular Disease and Mortality in U.S. Adults With Prediabetes: A Nationwide Cross‐Sectional and Prospective Cohort Study,” Acta Diabetologica 61, no 11 (2024): 1413–1421, 10.1007/s00592-024-02305-1. [DOI] [PubMed] [Google Scholar]
- 11. Kong X. and Wang W., “Prediabetes Phenotypes and All‐Cause or Cardiovascular Mortality: Evidence From a Population‐Based Study,” Endocrine Practice 31, no 4 (2025): 486–493, 10.1016/j.eprac.2025.01.003. [DOI] [PubMed] [Google Scholar]
- 12. Kong X., Gao X., and Wang W., “Oxidative Balance Score and Associations With Dyslipidemia and Mortality Among US Adults: A Mortality Follow‐Up Study of a Cross‐Sectional Cohort,” JPEN Journal of Parenteral and Enteral Nutrition 48, no 6 (2024): 735–745, 10.1002/jpen.2661. [DOI] [PubMed] [Google Scholar]
- 13. American Diabetes Association Professional Practice Committee , “2. Diagnosis and Classification of Diabetes: Standards of Care in Diabetes‐2024,” Diabetes Care 47, no Suppl 1 (2024): S20–S42, 10.2337/dc24-S002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Booth J. N. 3rd, Li J., Zhang L., Chen L., Muntner P., and Egan B., “Trends in Prehypertension and Hypertension Risk Factors in US Adults: 1999–2012,” Hypertension 70, no. 2 (2017): 275–284, 10.1161/HYPERTENSIONAHA.116.09004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Ma C., Zhai L., Huo R. R., and Liu Y. H., “Joint Associations of Pre‐diabetes, Pre‐hypertension, and Pre‐Dyslipidemia With Cardiovascular and Metabolic Disease Progression,” Diabetology & Metabolic Syndrome 17, no. 1 (2025): 310, 10.1186/s13098-025-01896-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Li Y., Kong X., and Wang W., “Sex Differences in the Associations Between Prior Weight Loss and All‐Cause or Cardiovascular Mortality in Non‐Elderly Individuals With Hyperuricemia: a Mortality Follow‐Up Study,” Nutrition & Metabolism 22, no. 1 (2025): 43, 10.1186/s12986-025-00930-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Levey A. S., Stevens L. A., Schmid C. H., et al., “A New Equation to Estimate Glomerular Filtration Rate,” Annals of Internal Medicine 150, no. 9 (2009): 604–612, 10.7326/0003-4819-150-9-200905050-00006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Mainous A. G. 3rd, Everett C. J., Liszka H., King D. E., and Egan B. M., “Prehypertension and Mortality in a Nationally Representative Cohort,” American Journal of Cardiology 94, no. 12 (2004): 1496–1500, 10.1016/j.amjcard.2004.08.026. [DOI] [PubMed] [Google Scholar]
- 19. Hozawa A., Kuriyama S., Kakizaki M., Ohmori‐Matsuda K., Ohkubo T., and Tsuji I., “Attributable Risk Fraction of Prehypertension on Cardiovascular Disease Mortality in the Japanese Population: The Ohsaki Study,” American Journal of Hypertension 22, no. 3 (2009): 267–272, 10.1038/ajh.2008.335. [DOI] [PubMed] [Google Scholar]
- 20. Lorenzo C., Aung K., Stern M. P., and Haffner S. M., “Pulse Pressure, Prehypertension, and Mortality: The San Antonio Heart Study,” American Journal of Hypertension 22, no. 11 (2009):1219–1226, 10.1038/ajh.2009.151. [DOI] [PubMed] [Google Scholar]
- 21. Huang Y. Q., Liu L., Huang J. Y., et al., “Prehypertension and Risk for All‐Cause and Cardiovascular Mortality by Diabetes Status: Results From the National Health and Nutrition Examination Surveys,” Annals of Translational Medicine 8, no. 6(2020): 323, 10.21037/atm.2020.02.144. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Schlesinger S., Neuenschwander M., Barbaresko J., et al., “Prediabetes and Risk of Mortality, Diabetes‐Related Complications and Comorbidities: Umbrella Review of Meta‐Analyses of Prospective Studies,” Diabetologia 65, no. 2 (2022): 275–285, 10.1007/s00125-021-05592-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Cao Z., Li W., Wen C. P., et al., “Risk of Death Associated With Reversion From Prediabetes to Normoglycemia and the Role of Modifiable Risk Factors,” JAMA Network Open 6, no. 3 (2023): e234989, 10.1001/jamanetworkopen.2023.4989. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Huang Y. Q., Liu L., Huang C., et al., “Impacts of Pre‐Diabetes or Prehypertension on Subsequent Occurrence of Cardiovascular and All‐Cause Mortality Among Population Without Cardiovascular Diseases,” Diabetes, Metabolic Syndrome and Obesity: Targets and Therapy 13, (2020): 1743–1752, 10.2147/DMSO.S255842. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Khosravi A., Gharipour M., Nezafati P., et al., “Pre‐Hypertension, Pre‐Diabetes or Both: Which is Best at Predicting Cardiovascular Events in the Long Term?” Journal of Human Hypertension 31, no. 6 (2017): 382–387, 10.1038/jhh.2016.42. [DOI] [PubMed] [Google Scholar]
- 26. Gupta A. K., Brashear M. M., and Johnson W. D., “Coexisting Prehypertension and Prediabetes in Healthy Adults: A Pathway for Accelerated Cardiovascular Events,” Hypertension Research 34, no. 4 (2011): 456–461, 10.1038/hr.2010.267. [DOI] [PubMed] [Google Scholar]
- 27. Cersosimo E. and DeFronzo R. A., “Insulin Resistance and Endothelial Dysfunction: The Road Map to Cardiovascular Diseases,” Diabetes Metabolism Research and Reviews 22, no. 6 (2006): 423–436, 10.1002/dmrr.634. [DOI] [PubMed] [Google Scholar]
- 28. Stahel P., Xiao C., Hegele R. A., and Lewis G. F., “The Atherogenic Dyslipidemia Complex and Novel Approaches to Cardiovascular Disease Prevention in Diabetes,” Canadian Journal of Cardiology 34, no. 5 (2018): 595–604, 10.1016/j.cjca.2017.12.007. [DOI] [PubMed] [Google Scholar]
- 29. Zhang L., Zhang Y., Shen S., et al., “Safety and Effectiveness of Metformin Plus Lifestyle Intervention Compared With Lifestyle Intervention Alone in Preventing Progression to Diabetes in a Chinese Population With Impaired Glucose Regulation: A Multicentre, Open‐Label, Randomised Controlled Trial,” Lancet Diabetes & Endocrinology 11, no. 8 (2023): 567–577, 10.1016/S2213-8587(23)00132-8. [DOI] [PubMed] [Google Scholar]
- 30. Yang H., Liu Y., Huang Z., and Deng G., “Achieving Prediabetes Reversal in China: A Nationwide Longitudinal Study on the Role of Blood Glucose and Lipid Management in Middle‐Aged and Elderly Adults,” Frontiers in Endocrinology 15 (2025): 1463650, 10.3389/fendo.2024.1463650. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Tables: jch70264‐sup‐0001‐tableS1‐S2.doc
Data Availability Statement
The dataset for this study is publicly available at CDC website at https://wwwn.cdc.gov/nchs/nhanes/Default.aspx