Association between coexisting hypertension, dyslipidaemia and elevated C reactive protein with cardiovascular disease and mortality: a cross-sectional and longitudinal analysis in a representative cohort of older US adults

Abstract

Objective

Hypertension and dyslipidaemia are established risk factors for cardiovascular disease (CVD) but they are often insufficient on their own to predict CVD. Inflammation also contributes to CVD, but research on the co-occurrence of inflammation, hypertension and dyslipidaemia and CVD risk is limited. Knowledge of inflammatory status in addition to other risk factors is vital for clinicians to correctly evaluate patients for CVD risk.

Methods

Prospective data from the Health and Retirement Study, a representative cohort of US adults over 50 years of age (n=7895), were used. The average participant age was 68.8 years, and 54.9% were female. 80.7% were non-Hispanic white, 10.1% were non-Hispanic black and 9.2% were Hispanic. Hypertension, dyslipidaemia and elevated C reactive protein (CRP) were used to create a CVD risk score: low (0–1 factors), medium (2 factors) or high (all 3 factors). Measurement and definition guidelines for these variables are thoroughly explained in the methods section. Weighted logistic regression models estimated the OR of (1) prevalent and incident CVD for medium and high-risk groups versus the low-risk group and (2) 4-year mortality adjusting for covariates.

Results

Cross-sectionally, high-risk participants (n=1706) had significantly higher odds of CVD prevalence compared with participants with low-risk (n=3107) (adjusted OR 1.54, 95% CI: (1.29 to 1.84)). Medium-risk (n=3082) participants had higher odds of CVD prevalence, though this did not reach significance. Prospectively, medium-risk and high-risk participants had significantly higher odds of 4-year CVD incidence (medium-risk adjusted OR 1.57, 95% CI (1.18 to 2.09); high-risk adjusted OR 1.67, 95% CI (1.19 to 2.36)) compared with those with low risk. Risk of 4-year mortality was higher in high-risk (OR 2.12, 95% CI (1.60 to 2.8)) participants versus low-risk, and non-significantly elevated in medium-risk participants.

Conclusions

Co-occurrence of hypertension, dyslipidaemia and elevated CRP was strongly associated with increased CVD prevalence, higher incident CVD and elevated 4-year mortality in older US adults, emphasising the importance of multifactor screening for CVD risk.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Hypertension, inflammation and hypercholesterolaemia are risk factors for cardiovascular disease (CVD) and related mortality.

• Individual assessments of these risk factors often do not fully explain CVD risk.

WHAT THIS STUDY ADDS

• Co-occurrence of hypertension, hypercholesterolaemia and systemic inflammation as measured by C reactive protein is associated with current CVD, 4-year CVD incidence and 4-year mortality.

• Occurrence of any of these conditions on their own is not associated with current or future CVD or 4-year mortality.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• Thorough assessments of blood pressure, lipids and inflammation in combination with one another are necessary to identify those at risk of CVD.

• Implementing treatments targeting all of these factors may be more helpful in reducing risk of CVD and related deaths.

Introduction

Despite being one of the most researched chronic health conditions, cardiovascular disease (CVD) is the leading cause of death in the USA and globally. CVD describes a range of conditions impacting the heart and vascular system, with the most common being stroke, ischaemic heart disease and congestive heart failure.^{1 2} CVD cases have nearly doubled from 271 million in 1990 to 523 million in 2019, and CVD-related deaths have also increased from 12.1 million in 1990 to 18.6 million in 2019.³ Aside from contributing to morbidity and mortality globally, CVD has placed an extreme burden on healthcare systems and societies. In 2010, the global cost of CVD was around US$865 billion and is estimated to rise to US$1.05 trillion in 2030. Around half of the monetary loss is due to direct healthcare costs and the other half is productivity loss from not working, disability or premature death.⁴ Therefore, the development of diagnostic tools for early identification of individuals at risk of developing CVD is crucial for timely diagnosis and intervention.

Multiple biological phenomena, including dyslipidaemia, inflammation and endothelial dysfunction, lead to the development of atherosclerosis, the main driver of CVD.⁵ Well-known factors contributing to these biomarkers are modifiable and include hypertension (HTN), dyslipidaemia, diabetes mellitus (DM), smoking, obesity, unhealthy diet and sedentary lifestyle.² In the USA, the number of adults with HTN is about 116 million and globally is about 3 billion.³ In the USA, around 38% of adults have total cholesterol (TC) ≥200 mg/dL, 28% of adults have low-density lipoprotein (LDL) ≥130 mg/dL, 21% of adults have triglyceride (TG)≥150 mg/dL and 17% of adults have high-density lipoprotein (HDL) <40 mg/dL.⁶ Both HTN and dyslipidaemia are indicated in the development of CVD, but studies have also demonstrated their insufficiency in independently predicting all cases of CVD.7,9 Inflammation has emerged as another key CVD risk factor. Inflammation is the immune system’s reaction to dangerous substances, such as pathogens, damaged cells or toxins. This process is crucial for maintaining health as it removes damaging entities and promotes healing. Acute inflammation works to mitigate injury or infection, but if inflammation becomes chronic, it can wreak havoc on the body and contribute to multiple debilitating health conditions, such as rheumatoid arthritis, lupus, periodontitis and atopic dermatitis.¹⁰ Research has shown a strong correlation between chronic inflammatory states and the risk of developing atherosclerotic plaque and CVD,¹¹ and to a degree, increased risk for all-cause mortality.¹² In blood vessels, inflammation occurs in response to vessel injury, oxidation of serum lipids and infection. HTN damages vessel endothelium, prompting an inflammatory response. In response to endothelial damage, leucocytes bind monocytes to the affected site. Dyslipidaemia increases the risk of serum lipids getting stuck behind the endothelium and becoming oxidised. This triggers a stronger inflammatory response because monocytes interact with these oxidised particles and are more likely to stay bound to the endothelium. Furthermore, monocytes can transform into macrophages and foam cells, which are precursors in the development of plaque. Other factors, including DM, smoking, obesity, unhealthy diet and sedentary lifestyle, amplify the harmful effects of inflammation, HTN and dyslipidaemia.13,15 Systemic inflammation can be measured by a variety of blood-based biomarkers, the most prominent being C reactive protein (CRP), which is independently associated with CVD.¹¹

Although prior research examines the link between CRP, HTN and CVD risk,¹⁶ as well as rates of co-occurrence of HTN, DM and elevated CRP,¹⁷ there is limited research on the co-occurrence of inflammation, HTN and dyslipidaemia and resultant CVD risk. In particular, it is not well understood if an individual needs all three factors to develop CVD. Additionally, no past study has looked at these three risk factors and CVD outcomes in a representative cohort of older US adults. In the present study, we are addressing this gap in the literature by determining whether having all three risk factors—high inflammation, HTN and dyslipidaemia—is correlated with a greater risk of prevalent and incident CVD diagnoses and mortality relative to 0–1 or 2 of these risk factors, using data from the Health and Retirement Study (HRS), a nationally representative, longitudinal study of adults over 50 years.

Methods

Study design and data sources

Data for this study came from the HRS, an ongoing nationally representative panel study that surveys around 20 000 people older than age 50 in the USA every 2 years. This study is supported by the National Institute on Aging and Social Security Administration and is coordinated by the University of Michigan Institute for Social Research. The design and methodology of HRS have been thoroughly outlined elsewhere.^{18 19} All data are publicly available at https://hrs.isr.umich.edu/data-products.²⁰ In this study, we first conducted a cross-sectional analysis examining the association between CVD prevalence and clinical risk factors for CVD using data from 2016 (wave 13). We then examined the association between clinical risk factors for CVD and CVD incidence and all-cause mortality using data from 2016 and 2020 (wave 15). Of the 9850 participants with biomarker data available in 2016, we excluded individuals missing data on blood pressure (BP), lipids and CRP, as well as covariates including age, sex, race/ethnicity, DM or hyperglycaemia, smoking history and obesity to obtain a final sample of 7895 participants (online supplemental figure 1). For longitudinal analysis of 4-year CVD incidence, 5620 participants in 2016 did not have self-reported CVD, and of those, only 4708 were available for analysis in 2020 due to attrition.

Patient and public involvement

Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

Exposure

Baseline CVD risk score was generated from three clinical conditions known to contribute to CVD: HTN, dyslipidaemia and elevated CRP. HTN status was obtained from HRS self-reported health data on the basis of a healthcare provider’s diagnosis. LDL-cholesterol (LDL-C) was calculated in serum specimens having a TG value <400 mg/dL using the formula of Friedewald LDL-C=TC−HDL-cholesterol (HDL-C)−TG/5.0. HDL-C was measured directly in serum using the Roche HDL-C third generation direct method (Roche Diagnostics, Indianapolis, Indiana, USA) on a Roche Cobas 6000 Chemistry Analyzer (Roche Diagnostics Corporation). TG was measured in serum using an enzymatic TG Reagent (Roche Diagnostics) on a Roche Cobas 6000 Chemistry Analyzer (Roche Diagnostics). Dyslipidaemia was categorised as LDL-C >130 mg/dL, TG >150 mg/dL or HDL-C <50 mg/dL for women and <40 mg/dL for men, based on clinical reference ranges.²¹ CRP was measured in serum using a latex-particle enhanced immunoturbidimetric assay kit (Roche Diagnostics) and read on the Roche COBAS 6000 Chemistry analyzer (Roche Diagnostics). High CRP was categorised as >3 mg/L, based on the clinical reference range.^{22 23} Values of continuous CRP were log transformed for descriptive statistics.

Low risk was defined as having 0–1 diagnoses of HTN, high CRP or dyslipidaemia, medium risk was defined as having 2 of these factors and high-risk was defined as having all 3 factors.

Outcome

The outcomes of interest for this study were prevalent CVD, 4-year incident CVD and 4-year incident mortality. CVD diagnosis was determined through self-reported health data of a healthcare provider’s diagnosis of heart attack, coronary heart disease, angina, congestive heart failure or other heart problems, and stroke or transient ischaemic attack (TIA). Prevalent CVD was defined as presence of a heart condition or a stroke or TIA at baseline per participant report. Incident CVD was defined as the first occurrence of a heart condition or a stroke or TIA during the 4-year follow-up period from 2016 to 2020. Mortality data were collected through interviews with surviving participants or their next-of-kin and searches in the National Death Index. A 4-year incident mortality was defined as death from any cause occurring during the 4-year follow-up period.

Covariates

Covariates included age (years), sex assigned at birth (female/male), race/ethnicity (non-Hispanic white, non-Hispanic black or Hispanic), obesity (yes/no), ever versus never smoker (yes/no), DM or hyperglycaemia (yes/no). Age, sex and race/ethnicity were obtained from HRS self-reported demographic data. The presence of DM or hyperglycaemia was obtained from HRS self-reported health data on the basis of a healthcare provider’s diagnosis. Ever versus never smoker status was obtained from HRS self-reported health data. A participant was classified as obese if their calculated body mass index was ≥30 kg/m², based on the clinical classification of obesity.²⁴

Statistical analysis

Continuous variables were reported as weighted mean±SE, and categorical variables were reported as frequency (weighted percentage). Distributions of covariates by CVD diagnosis and risk score were reported using F-test for continuous variables and Rao-Scott χ² test for categorical variables. Weighted logistic regression models were used to calculate adjusted ORs of CVD or mortality in participants in the medium and high-risk groups compared with the low-risk group. Model 1 was adjusted for age. Model 2 was adjusted for age, sex, race, obesity, ever-smoker status and DM. The mortality analysis additionally adjusted for self-reported CVD in 2016.

All analyses were completed using SAS V.9.4 (SAS Institute) using Proc Survey procedures. P values at alpha <0.05 were considered statistically significant.

Results

Baseline characteristics

Descriptive statistics for those with and without CVD at baseline are provided in table 1. The mean age of participants with CVD was 72.25±0.27 years and 67.39±0.15 years for participants without CVD. Of the participants with CVD, 50.17% were female, and of the participants without CVD, 56.55% were female. There was a higher proportion of non-Hispanic whites in the CVD group, a greater proportion of Hispanics in the non-CVD group and similar proportions of non-Hispanic blacks in the two groups. There was a greater prevalence of obesity as well as DM or hyperglycaemia in the CVD group, while the percentage of ever-smokers was not significantly different between the two groups. Mean LDL-C and HDL-C levels were significantly greater in the non-CVD group, mean log CRP was significantly greater in the CVD group and mean TG was similar between the two groups. HTN and high CRP were more prevalent in the CVD group, while dyslipidaemia was more prevalent in the non-CVD group.

Table 1. Characteristics of HRS participants with and without CVD in 2016 (n=7895).

Variables	CVD (n=2275)	No CVD (n=5620)	P value
Age (years)	72.25±0.27	67.39±0.15	<0.0001
Sex (female), n (%)	1215 (50.17)	3424 (56.55)	<0.0001
Race/ethnicity, n (%)
Non-Hispanic white	1678 (82.03)	3741 (80.26)	0.0482
Non-Hispanic black	371 (10.30)	1037 (10.05)
Hispanic	239 (7.67)	870 (9.68)
Obesity, n (%)	880 (36.56)	2000 (32.18)	0.0024
Ever smoker, n (%)	219 (10.39)	614 (11.68)	0.2120
DM or hyperglycaemia, n (%)	861 (35.15)	1330 (19.94)	<0.0001
LDL-C (mg/dL)	90.49±0.96	107.46±0.58	<0.0001
HDL-C (mg/dL)	53.72±0.47	60.02±0.34	<0.0001
TG (mg/dL)	140.49±1.90	136.30±1.14	0.0542
Log CRP	0.79±0.05	0.60±0.03	0.0007
HTN, n (%)	1800 (75.85)	3319 (52.93)	<0.0001
Dyslipidaemia, n (%)	1450 (64.04)	4132 (73.86)	<0.0001
High CRP, n (%)	993 (41.30)	2210 (36.51)	0.0014

Values are mean±standard errorSE and frequency (percentage); Bold=statistically significant at alpha <0.05.

CRP, C reactive protein; CVDcardiovascular diseaseDMdiabetes mellitusHDL-C, high-density lipoprotein cholesterol; HRSHealth and Retirement StudyHTN, hypertensionLDL-C, low-density lipoprotein cholesterol; TG, triglyceride

Distributions of covariates for low, medium and high-risk factor scores based on presence of 0–1, 2 or 3 of HTN, dyslipidaemia and high CRP, respectively, are described in table 2. Low-risk group mean age was 68.31±0.20 years, medium-risk group mean age was 69.12±0.22 years and high-risk group mean age was 68.64±0.31 years. 51.01%, 56.84% and 59.61% of participants in low, medium and high-risk groups were female, respectively. The proportion of non-Hispanic whites was highest in the low-risk group, while the proportion of non-Hispanic blacks was highest in the high-risk group. Both medium and high-risk groups contained a similarly greater proportion of Hispanic participants compared with the low-risk group. There was an increasing prevalence of obesity, ever-smokers and DM or hyperglycaemia as the number of risk factors increased. Mean LDL-C, TG and log CRP increased with increasing risk factor scores, while mean HDL-C decreased.

Table 2. Characteristics of HRS participants by risk category in 2016 (n=7895).

Variables	Low risk (n=3107)	Medium risk (n=3082)	High risk (n=1706)	P value
Age (years)	68.31±0.20	69.12±0.22	68.64±0.31	0.0255
Sex (female), n (%)	1696 (51.01)	1844 (56.84)	1099 (59.61)	<0.0001
Race/ethnicity, n (%)
Non-Hispanic white	2356 (85.68)	2057 (78.68)	1006 (73.72)	<0.0001
Non-Hispanic black	415 (6.98)	557 (10.82)	436 (15.69)
Hispanic	354 (7.34)	481 (10.50)	274 (10.59)
Obesity, n (%)	696 (19.62)	1177 (36.50)	1007 (57.50)	<0.0001
Ever smoker, n (%)	293 (10.09)	322 (11.48)	218 (13.85)	0.0136
DM or hyperglycaemia, n (%)	623 (16.46)	894 (26.16)	674 (36.27)	<0.0001
LDL-C (mg/dL)	97.53±0.75	105.99±0.80	109.27±1.24	<0.0001
HDL-C (mg/dL)	64.10±0.45	56.03±0.43	50.14±0.52	<0.0001
TG (mg/dL)	112.84±1.20	148.94±1.61	169.61±2.32	<0.0001
Log CRP	0.09±0.03	0.78±0.04	1.59±0.04	<0.0001
HTN, n (%)	1028 (27.89)	2375 (73.45)	1716 (100.00)
Dyslipidaemia, n (%)	1346 (46.90)	2520 (84.40)	1716 (100.00)
High CRP, n (%)	192 (5.70)	1295 (42.15)	1716 (100.00)

Values are mean±standard errorSE and frequency (percentage); Bold=statistically significant at alpha <0.05.

CRP, C reactive protein; DMdiabetes mellitusHDL-C, high-density lipoprotein cholesterol; HRSHealth and Retirement StudyHTN, hypertensionLDL-C, low-density lipoprotein cholesterol; TG, triglyceride

Cross-sectional analyses

Associations between risk factor score and 2016 CVD prevalence are shown in figure 1. After adjusting for age, the odds of CVD prevalence were 1.20-fold higher in medium-risk participants and 1.76-fold higher in high-risk participants compared with low-risk participants (medium-risk 95% CI (1.04 to 1.38), high-risk 95% CI (1.49 to 2.08)). This association diminished in the medium-risk group following additional adjustment for sex, race/ethnicity, obesity status, ever-smoker status and DM or hyperglycaemia status (OR 1.13, 95% CI: (0.97 to 1.31)) but remained robust in the high-risk group (OR 1.54, 95% CI (1.29 to 1.84)).

Figure 1. Association (ORs and 95% CIs) between risk category and CVD prevalence in 2016 in HRS participants (n=7895). CVD, cardiovascular disease; HRS, Health and Retirement Study.

Longitudinal analyses

Incident CVD ORs based on risk factor score are depicted in figure 2. After full adjustment, a 4-year CVD incidence risk among medium and high-risk participants was significantly greater compared with low-risk participants, with OR values of 1.57 (95% CI (1.18 to 2.09)) and 1.67 (95% CI (1.19 to 2.36)) for medium and high-risk participants, respectively. Incident mortality ORs based on risk factor score are depicted in figure 2. With full covariate adjustment, the risk of 4-year incident mortality was statistically significantly elevated in high-risk participants (OR 2.12, 95% CI (1.60 to 2.80)). A 4-year incident mortality risk was also increased in medium-risk participants, though this was not statistically significant (OR 1.21, 95% CI (0.97 to 1.51)).

Figure 2. Association (ORs and 95% CIs) between risk category and CVD incidence between 2016 and 2020 (n=4708) and 4-year mortality (n=7895) in HRS participants. CVD, cardiovascular disease; HRS, Health and Retirement Study.

Sensitivity analyses

We compared the presence of all three risk factors (HTN, dyslipidaemia and high CRP) versus only having HTN and dyslipidaemia in our fully adjusted model. We found that those will all three risk factors had higher odds of prevalent CVD in 2016 (OR 1.31, 95% CI 1.08, 1.58) and higher odds of 4-year all-cause mortality (OR 2.27, 95% CI 1.68, 3.05) than those participants with only HTN and dyslipidaemia. Additionally, we examined the association between high and medium CVD risk versus low risk for the two CVD outcomes that were combined for the main analyses (1) self-reported heart problems (healthcare provider’s diagnosis of heart attack, coronary heart disease, angina, congestive heart failure or other heart problem) and (2) self-reported stroke/TIA in 2016 separately. We observed a statistically significant association between high risk and heart problems in 2016 (OR 1.54, 95% CI 1.29, 1.84) as well as stroke prevalence in 2016 (OR 1.50, 95% CI 1.07, 2.11). No statistically significant associations were observed for medium versus low risk and these individual outcomes.

Discussion

HTN and dyslipidaemia are well established as contributing factors in the development of CVD. More recently, elevated inflammatory status, measured through CRP, has been explored as an important element in the aetiology of CVD. With growing evidence supporting the synergistic effects of HTN, dyslipidaemia and inflammation on CVD risk, we sought to investigate the prevalence and incidence of CVD and mortality between groups with 0–1, 2 or 3 of these conditions in a nationally representative sample of US adults over the age of 50.

Our findings revealed that those with 0–1 or 2 risk factors had similar prevalence of CVD, but those with all 3 CVD risk factors had a significantly higher prevalence of CVD, even after adjusting for age, sex, race/ethnicity, obesity status, ever-smoker status, and DM or hyperglycaemia. Additionally, we found that having an increased number of our proposed CVD risk factors was strongly associated with incident CVD, even after adjusting for multiple confounding variables and observed elevated odds of incident mortality with all three risk factors. These findings support the hypothesis that CVD is multifactorial and screening for all three factors may be clinically important for identifying and managing individuals at high risk of developing CVD, as well as those at risk of premature death.

These results are consistent with previous research, including multiple observational and experimental studies. Initially, based on the cholesterol hypothesis of CVD, which highlights cholesterol as the primary driver of atherosclerosis, various medications were developed to reduce cholesterol levels. The most promising were niacin and torcetrapib, as both produced significant improvements in lipid profiles. Despite these improvements, participants demonstrated zero clinical benefit, and some even had worse CVD and mortality outcomes.^{25 26} Additionally, even on multidrug lipid-lowering therapy, 20% of participants experienced residual CVD risk.²⁷ This residual CVD risk was associated with CRP elevation, despite achieving control of cholesterol.²⁸ Other CVD drug trials have found that the anti-inflammatory effects of drugs, including statins, produce a protective effect against CVD and CVD-related deaths.²⁹ Our current knowledge has moved from believing that CVD stems from cholesterol buildup alone to a multifaceted disease. HTN has also been a dominant player in the development of CVD and inflammation has proven itself as well. Inflammation is involved in every step of atherosclerosis, and determining CVD risk based on multiple factors, not just BP and cholesterol, is prudent in the mitigation of CVD-related deaths. In practice, proper and comprehensive assessment of BP, lipids and CRP is rare, which is problematic in the prevention and management of CVD, a multifactorial condition. In this context, it is important to ensure the implementation of current diagnostic guidelines, which advocate for concurrent and routine BP, lipid and CRP measurements in addition to other important factors such as smoking status, obesity, statin usage and family history. Knowledge of all health parameters is vital to accurately describe CVD risk.

In the present study, participants with CVD tended to be older, more likely to be male, obese and have DM or hyperglycaemia, compared with those with no CVD, as expected. Additionally, participants with CVD had lower LDL-C and HDL-C levels, were more likely to have HTN and elevated CRP and were less likely to be dyslipidaemic. The lower prevalence of dyslipidaemia in those with CVD could be due to lipid-lowering medication use, which is a first-line agent in the prevention of future CVD events in those previously diagnosed with CVD. A thorough analysis of medication usage is beyond the scope of this paper but is an area for future research. We noted a trend of a higher percentage of females as the number of risk factors increased. We also found multiple differences in distribution of racial and ethnic groups between risk categories, with the ratio of non-Hispanic whites decreasing as the number of risk factors increased, the ratio of non-Hispanic blacks increasing as the number of risk factors increased, and Hispanics representing a greater proportion of participants in both medium and high-risk groups. Additionally, high-risk participants were more likely to be obese, have a history of smoking and have DM or hyperglycaemia.

This study has many strengths. Data come from the HRS, and the results are generalisable to non-institutionalised US population over age 51. Measures of key laboratory variables, including CRP and lipids, were obtained using a standard protocol. HRS contains a wealth of information on covariates, allowing the analyses to be controlled for potential confounders of the association. Additionally, the longitudinal design provides insight into the temporal order of CVD risk factors and CVD incidence, lending insight into causation. Limitations of this study warrant consideration. Covariate information, including CVD, obesity, smoking history, DM or hyperglycaemia and HTN, was all self-reported by participants, and these data can be influenced by memory, social desirability and the absence of diagnosis. Ever-smoker status and obesity are highly variable across individuals, and with a yes/no metric, it is difficult to ascertain the severity of these variables. In the future, this study should be replicated with objective and quantifiable health measurements, such as laboratory tests or physical measurements taken by trained professionals, which is not possible in large, nationally representative surveys. Additionally, future studies could consider using more detailed measures for smoking history and obesity, such as pack-years or body composition analysis, to provide a more nuanced understanding of these variables and their associations with CVD. Other lifestyle information, such as physical activity, diet and stress, which are all independent factors in development of CVD,30,32 were not included as covariates. Furthermore, pharmacological interventions were not analysed, as it was outside the scope of this study. Finally, biomarkers shown to more accurately predict CVD risk, such as LDL particle size, lipoprotein(a) and apolipoprotein B, were not available for analysis in HRS.^{33 34} Given the paucity of data regarding usage rates of CRP in the clinical setting, it is advisable to examine both the utilisation rates of CVD risk calculators and the integration of CRP measurements in clinical settings. Future studies should also focus on implementing a variety of lifestyle factors and above-mentioned biomarkers on top of the CVD risk factors examined in this study to develop a comprehensive score of CVD risk.

Conclusions

Co-occurrence of HTN, dyslipidaemia and elevated CRP was strongly associated with increased CVD prevalence, higher incident CVD and elevated 4-year mortality in older US adults. Our findings emphasise the importance of multifactor screening for assessing CVD risk in the clinical setting.

Data availability statement

Data are available in a public, open access repository.