Abstract
Background
Cardiovascular disease (CVD) often coexists with dementia as the contributor to disease burden in the aging population. Although the effects of several modifiable factors on the CVD and dementia risk have been increasingly recognized, their impacts on their dynamic transitions have not yet been explored. Therefore, we aimed to explore the role of Life’s Crucial 9 (LC9; which includes measurements for diet, physical activity, smoking, BMI, cholesterol, blood pressure, blood sugar, sleep, and psychological health) in the dynamic transitions of CVD and dementia.
Methods
Using the UK Biobank (351,945 CVD-free participants, aged 40–70, and 55.1% female), we applied Cox proportional hazard model and the multi-state model to explore the relationships of LC9 with the dynamic transitions of CVD to dementia. We further examined which component of LC9 (behavior subscale, biological subscale, and psychological health) contributed to these dynamic transitions.
Results
Of the total CVD-free participants, we find that there are 35,316 newly onset CVD cases and 5426 dementia cases with a maximum follow-up of 16 years. Findings reveal that higher LC9 scores are associated with a lower risk of CVD, dementia, and dynamic transitions from CVD to dementia. Furthermore, these results suggest that psychological health contributes more strongly to dynamic transitions from CVD to dementia than the behavioral or biological subscales (Atrial fibrillation to dementia: hazard ratio [HR] = 1.48, 95% confidence interval [CI] = 1.20–1.81; Ischemic stroke to dementia: HR = 1.39, 95% CI = 1.02–1.89; Myocardial infarction to dementia: HR = 1.48, 95% CI = 1.03–2.13; Heart failure to dementia: HR = 1.59, 95% CI = 1.22–2.07).
Conclusions
Our findings highlight the need for closer surveillance of psychological health in CVD patients to lower the risk of subsequent dementia incidence.
Subject terms: Cardiovascular diseases, Lifestyle modification, Dementia
Zhang et al. explore the role of Life’s Crucial 9 (Life’s Essential 8 plus psychological health) in the dynamic transitions of cardiovascular disease (CVD) and dementia in the UK Biobank. The findings highlight the need for closer surveillance of psychological health in CVD patients to lower the risk of subsequent dementia incidence.
Plain language summary
Although the effects of lifestyle factors on the developing cardiovascular disease (CVD) and dementia have been increasingly recognized, their impacts on the dynamic transitions of CVD and dementia remains underexplored. Here, we studied the role of Life’s Crucial 9 (LC9; which includes measurements for diet, physical activity, smoking, BMI, cholesterol, blood pressure, blood sugar, sleep, and psychological health) in the transitions from CVD to dementia using a large UK database. We find that better LC9-defined cardiovascular health is associated with lower risks of dynamic transitions of CVD and dementia. We also find that psychological health contributes more strongly to all transitions from CVD to dementia. Our findings highlight the role of lifestyle behaviors, especially psychological health, in CVD patients to lower the risk of subsequent dementia incidence.
Introduction
Cardiovascular disease (CVD) is one of the leading causes of morbidity, disability, and mortality worldwide1. Notably, CVD often coexists with dementia as the major contributor to disease burden in the aging population. Stroke has been identified as a risk factor for dementia. Moreover, atrial fibrillation (AF) is associated with several adverse comorbidities, including the development of dementia in patients with or without a diagnosis of stroke2–5. In addition, previous studies found that patients who survive a myocardial infarction (MI) are at higher risk for chronic heart failure (HF)6, and are more likely to develop dementia7. At the same time, the risk of subsequent cognitive impairment and dementia appeared to be higher in patients with HF8. Thus, it is reasonable to hypothesize the following disease trajectories 1) from event-free to incident AF, ischemic stroke (IS), and dementia; 2) from event-free to incident MI, HF, and dementia; 3) from event-free to incident AF, HF, and dementia; 4) from event-free to incident AF, MI, and dementia.
The role of psychological health on the risk of CVD and dementia have received increasing attention. Accumulating studies have demonstrated that depression is associated with elevated risks of CVD and dementia9,10. In a recent perspective, they introduced a new metric called Life’s Crucial 9 (LC9), including diet, physical activity, smoking, body mass index (BMI), blood lipids, blood pressure, blood sugar, sleep, and psychological health11. LC9 incorporates psychological health as a new component into Life’s Essential 8 (LE8) to improve cardiovascular health11. However, the agreement on how to combine psychological health factors is still lacking since the multidimensionality of psychological health11,12, and it remains unclear to what extent LC9 with psychological health could improve cardiovascular health. Moreover, considering many risk factors shared between CVD and dementia, we hypothesized that LC9 might be also associated with the long-term risk of dementia13. However, supportive evidence is also lacking.
Despite the known effects of some lifestyle factors or biological indicators on the risk of CVD or dementia, it remains unknown whether these modifiable factors have different relationships with the dynamic transitions of CVD and dementia, such as from event-free to AF incidence, subsequently to IS, and further to dementia. Previous studies have concentrated on the impacts of lifestyle factors or air pollution on progression from healthy to first cardiometabolic disease, cardiometabolic multimorbidity, and death14,15; however, evidence regarding the associations between LC9 and dynamic transitions of CVD and dementia is lacking.
In this study, we aim to examine the associations of combined and individual LC9 with the risk of CVD and dementia based on data from the UK Biobank. Importantly, the multi-state model is applied to investigate the potentially different relationships of LC9 with the dynamic transitions of CVD and dementia, and further identify which component of LC9 might be the major contributor to these dynamic transitions. In addition, the role of LC9 on the risk of dementia is further assessed in the CVD-prevalent cohort. We find that LC9 play an important role in the development of CVD or dementia, and dynamic transitions of CVD and dementia. Notably, psychological health disorders contribute more strongly to dynamic transitions from CVD to dementia. We conclude that closer surveillance of psychological health is needed in CVD patients to lower the risk of subsequent dementia.
Methods
Study population
The UK Biobank is a population-scale cohort study that recruited over 500,000 people between the ages of 40 and 70 years from 2006 to 2010. At baseline, self-administered touchscreen questionnaires were applied to collect information on participants’ sociodemographics, lifestyle exposures, and medical history. The participants also underwent physical measurements and provided blood samples.
The UK Biobank was approved by the National Information Governance Board for Health and Social Care and the National Health Service North West Centre for Research Ethics Committee (Ref: 11/NW/0382). All participants gave informed consent to participate and were followed up by linkage to the electronic health records of the UK National Health Service (NHS). Data from the UK Biobank are available to all researchers upon application (http://www.ukbiobank.ac.uk/). This study was conducted under the UK Biobank Application 90492 using anonymized data, and additional ethical review does not require according to the policies of our institute. Participant selection for the study is depicted in Supplementary Fig. 1.
Definition of Life’s Crucial 9
A modified version of the LC9 score was developed in this study (Supplementary Data 1). Briefly, LC9 is a mix of behavioral (healthy diet, physical activity, avoidance of tobacco, healthy sleep, and healthy weight), biological (healthy levels of blood lipids, blood glucose, and blood pressure)16,17, and psychological health targets. The information regarding the frequency of consumption of food items, physical activity, tobacco exposure, sleep duration, and medication use was collected by touchscreen questionnaires at baseline. The touchscreen questionnaire was later validated by the 24-h dietary recall in a subset of the UK Biobank participants, and has shown adequate agreement for each food group18. The details of the healthy diet scoring algorithm are shown in Supplementary Data 219. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured twice, and averaged values were used for analyses. Psychological health was defined by multiple factors, including depression, anxiety, chronic and traumatic stress, and social integration11, and their definitions are detailed in Supplementary Methods (Definition of Psychological Health). Briefly, the higher psychological scores indicate better psychological health. The psychological score was assigned as 100, 75, 50, 25, and 0 corresponding to having no psychological disorder, having one factor, having two factors, having three factors, and having four factors, respectively.
Each component ranged from 0 to 100 points, and the aggregate score was also scaled from 0 to 100 points based on the unweighted average of all 9 component scores. The following descriptive categories were proposed referring to the American Heart Association (AHA) advisory: high LC9 score (80–100), moderate LC9 score (50–79), and low LC9 score (0–49). We also calculated a behavior subscale based on the behavioral metrics (diet, physical activity, tobacco exposure, sleep health, and healthy weight), and a biopsychosocial subscale based on the biological metrics (blood lipids, blood glucose, and blood pressure) and psychological health.
Outcomes ascertainment
The diagnosis and incidence date of CVD and dementia were identified through hospital admissions, primary care, and self-reported data with the International Classification of Diseases Ninth and Tenth Revision (ICD-9, ICD-10) code. Code lists for outcomes are detailed in Supplementary Methods (Outcomes ascertainment), Supplementary Tables 1–3. The date and cause of death were obtained through linkage to the NHS Information Center and NHS Central Register. All participants were followed up from the date of enrollment and continued until the first incidence of interested outcomes (CVD or dementia), death, loss to follow-up, or the end of the study (December 31, 2021), whichever came first.
Covariates
Potential confounders included age at enrollment, sex, ethnicity, Townsend deprivation index (based on an individual’s postcode, a composite measure of deprivation based on unemployment, non-car ownership, non-home ownership, and household overcrowding), education, and current drinking. Details of covariates definitions are presented in Supplementary Table 4. Missing data on ethnicity (N = 1215; 0.3% of all participants), Townsend deprivation index (N = 429; 0.1% of all participants), education (N = 2666; 0.8% of all participants), and current drinking (N = 337; 0.1% of all participants) were recorded. For these covariates with missing data, we created an additional category to deal with their missing values.
Statistics and reproducibility
Baseline characteristics were presented as the mean ± standard deviation (SD) or median and interquartile range (IQR) for continuous variables, and as frequency and percentage for categorical variables.
We first used traditional Cox proportional hazard models with attained age as the underlying timescale to investigate the associations of combined and individual LC9 scores (behavior subscale, biological subscale, and psychological health) with the risk of MI, HF, AF, IS, and dementia. The models were adjusted for age, sex, ethnicity, education, Townsend deprivation index, and current drinking. Absolute risk (AR) was calculated to estimate the percentage of developments of CVD and dementia that theoretically would not have occurred if all participants were in the high LC9 category among the UK Biobank population. The Fine and Gray model was also conducted considering the potential competing risk of death. In addition, the exposure-response relationships of the LC9 score with the risk of MI, HF, AF, IS, and dementia were depicted using restricted cubic splines that were fitted based on Cox regression models with 4 knots at the quartiles of the LC9 score.
Subsequently, the multi-state survival models were further applied to assess the role of both combined and individual LC9 in the dynamic transitions of CVD and dementia in the CVD-free cohort. In the models, attained age was used as the underlying timescale, and the model was adjusted for age, sex, ethnicity, education, Townsend deprivation index, and current drinking. Multi-state model is an extension of the competing risk model by incorporating multiple subsequent or competing events as states of transitions to simultaneously assess the associations of certain factors with different stages of disease progression20. The multi-state models in our study were divided into four dynamic disease trajectories, and each trajectory consisted of four states including 1) baseline, AF, IS, and dementia, 2) baseline, MI, HF, and dementia, 3) baseline, AF, HF, and dementia, 4) baseline, AF, MI, and dementia. IS, HF, or MI was treated as the absorbing state, respectively (i.e., AF diagnosed after an IS, HF, or MI diagnosis was not considered, and MI diagnosed after a HF diagnosis was not considered). Accordingly, all participants started in the event-free state, and possible trajectories for each participant include (take one of the pathways as an example): 1) from baseline to AF, 2) from baseline to IS, 3) from baseline to dementia, 4) from AF to IS, 5) from AF to dementia, 6) from IS to dementia. Meanwhile, sensitivity analysis to reduce reverse causality was conducted by repeating the analyses after excluding CVD and dementia that occurred during the first two years of follow-up. Finally, in the CVD-prevalent cohort, Cox proportional hazard models were further conducted to validate the association of LC9 score with the risk of dementia among CVD patients.
All P values were two-sided, and P values < 0.05 were considered to indicate statistical significance. All statistical analyses were performed using Stata 15.1 and R 3.6.1.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Results
Baseline characteristics
The baseline characteristics of participants in the CVD-free and CVD-prevalent cohort according to low, moderate, and high LC9 scores are presented in Table 1 and Supplementary Table 5. Overall, participants with higher LC9 scores were more likely to be female, have higher education, and less likely to be socioeconomically deprived in both CVD-free and CVD-prevalent cohort. In addition, compared to participants in the CVD-free cohort, participants with a CVD diagnosis before cohort entry were older, more likely to be male, diabetic, hypertensive, and have psychological health disorders. The distribution of combined and individual component of LC9 score in the CVD-free and CVD-prevalent cohort are shown in Supplementary Figs. 2, 3.
Table 1.
Baseline characteristics of participants without CVD and dementia diagnosis before cohort entry according to the categories of Life’s Crucial 9 score
| Characteristics | Total (N = 351,945) | Life’s Crucial 9 score | ||
|---|---|---|---|---|
| Low (N = 20,654) | Moderate (N = 275,696) | High (N = 55,595) | ||
| Age at enrollment, y | 56.7 ± 8.1 | 57.1 ± 7.6 | 57.2 ± 8.0 | 54.2 ± 8.4 |
| Women | 193,983 (55.1) | 9153 (44.3) | 146,002 (53.0) | 38,828 (69.8) |
| White | 335,529 (95.3) | 19,432 (94.1) | 262,687 (95.3) | 53,410 (96.1) |
| Higher education | 173,011 (49.2) | 7661 (37.1) | 132,606 (48.1) | 32,744 (58.9) |
| Townsend deprivation index (Quintile 5) | 70,303 (20.0) | 6679 (32.3) | 54,607 (19.8) | 9017 (16.2) |
| Current drinking | 249,399 (70.9) | 13,217 (64.0) | 197,084 (71.5) | 39,098 (70.3) |
| Current smoking | 35,487 (10.1) | 7589 (36.7) | 27,399 (9.9) | 499 (0.9) |
| BMI, kg/m2 | 27.3 ± 4.7 | 32.5 ± 5.6 | 27.6 ± 4.4 | 23.9 ± 2.8 |
| Non-HDL, mg/dL | 165.8 ± 41.1 | 186.8 ± 45.7 | 169.2 ± 40.4 | 141.0 ± 31.6 |
| HbA1c, % | 5.4 ± 0.6 | 6.0 ± 1.1 | 5.4 ± 0.5 | 5.2 ± 0.3 |
| Diabetes | 18,321 (5.2) | 4598 (22.3) | 13,425 (4.9) | 298 (0.5) |
| Sleep duration, h/d | 7.2 ± 1.1 | 6.8 ± 1.6 | 7.1 ± 1.1 | 7.3 ± 0.8 |
| Moderate physical activity, min/wk | 100.0 (20.0, 300.0) | 0.0 (0.0, 70.0) | 90.0 (20, 280.0) | 180.0 (90.0, 420.0) |
| Vigorous physical activity, min/wk | 20.0 (0.0, 120.0) | 0.0 (0.0, 20.0) | 20.0 (0.0, 90.0) | 90.0 (0.0, 180.0) |
| SBP, mmHg | 137.8 ± 18.6 | 146.5 ± 17.1 | 139.6 ± 18.1 | 125.6 ± 16.3 |
| DBP, mmHg | 82.3 ± 10.1 | 88.4 ± 9.9 | 83.2 ± 9.8 | 75.6 ± 8.5 |
| Use of antihypertension medications | 64,004 (18.2) | 7612 (36.9) | 53,316 (19.3) | 3076 (5.5) |
| Psychological health disorders | 66,299 (18.8) | 8120 (39.3) | 52,466 (19.0) | 5713 (10.3) |
Values are mean ± SD, n (%), or median (IQR).
BMI body mass index, Non-HDL non-high-density lipoprotein cholesterol, HbA1c glycated hemoglobin, SBP systolic blood pressure, DBP diastolic blood pressure
Associations of LC9 with incident CVD and dementia among CVD-free participants
Among 351,945 CVD-free participants, 35,316 newly onset of CVD cases were recorded (during a median follow-up of 12.7 years), and 5426 participants were diagnosed with dementia (during a median follow-up of 14 years). Specifically, of all incident AF patients, 930 (5.1%) developed IS, 2082 (11.4%) developed HF, 1356 (7.6%) developed MI. Subsequently, 226 (3.9%) participants with IS, 236 (3.4%) participants with HF, and 241 (2.1%) participants with MI further developed dementia. In addition, 1256 (11.8%) participants with MI progressed to HF, and 276 (3.4%) participants with HF were further diagnosed with dementia (pathway baseline→ MI → HF→ dementia) (Fig. 2, Supplementary Tables 6–9).
Fig. 2. The associations between Life’s Crucial 9 score and dynamic transitions of cardiovascular diseases to dementia.
Left panel (a, c, e, and g), sample sizes and proportions of specific transitions are provided on the edge; Right panel (b, d, f, and h), the role of Life’s Crucial 9 score on the risk for transition from baseline to cardiovascular diseases and subsequent dementia. The multi-state survival models adjusted for age, sex, ethnicity, education, Townsend deprivation index, and current drinking. The hazard ratios (95% confidence intervals) for the associations between high Life’s Crucial 9 and specific transition are given on the edge. Statistically significant results are bolded.
Assessing by traditional Cox models and the competing risk models, higher LC9 scores were associated with a lower risk of MI, HF, AF, IS, and dementia, among CVD-free individuals (Supplementary Tables 10, 11). Behavior subscale scores (diet, physical activity, tobacco exposure, sleep health, and healthy weight) showed the strongest association with the risk of HF (hazard ratio [HR] = 0.71, 95% confidence interval [CI] = 0.69–0.72), while biological subscale scores (blood lipids, blood glucose, and blood pressure) showed the strongest association with the risk of MI (HR = 0.67, 95% CI = 0.66–0.69). Psychological health disorders showed the strongest association with the risk of dementia (HR = 1.48, 95% CI = 1.39–1.58). In addition, the exposure-response curves for the associations between LC9 score and the risk of MI, HF, AF, IS, and dementia are depicted in Fig. 1. Overall, linear trends for LC9 score associated with these conditions were observed, except for MI and dementia (P for non-linearity < 0.001).
Fig. 1. The associations between Life’s Crucial 9 score and the risk of myocardial infarction, heart failure, atrial fibrillation, ischemic stroke, and dementia incidence using restricted cubic spline regression (N = 351,945).
a for myocardial infarction, b for heart failure, c for atrial fibrillation, d for ischemic stroke, e for dementia. The model adjusted for age, sex, ethnicity, education, Townsend deprivation index, and current drinking. Solid red lines indicate hazard ratios; the shaded area represents the 95% confidence intervals; dotted black lines indicate reference line at HR = 1.
LC9 in the dynamic transitions from baseline to incident CVD and further to dementia
Furthermore, the multi-state models distinguished the role of LC9 in the dynamic transitions from baseline to CVD incidence and further to dementia. Briefly, combined LC9 scores were associated with all possible transitions, except for the transition from MI or HF to dementia (Fig. 2, Supplementary Tables 12–15). Besides, we found that the biopsychosocial subscale might contribute more strongly to all dynamic transitions from CVD to dementia than the behavior subscale (Supplementary Tables 16–19). After dividing biopsychosocial subscale into biological subscale and psychological health disorders, biological subscale was more likely to be associated with the transition from AF to IS (HR = 0.85, 95% CI = 0.79–0.91), while psychological health disorders mainly contributed to the transition from AF to dementia (pathway baseline→ AF → IS→ dementia: HR = 1.48, 95% CI = 1.20–1.81), the transition from IS to dementia (pathway baseline→ AF → IS→ dementia: HR = 1.39, 95% CI = 1.02–1.89), the transition from MI to dementia (pathway baseline→ MI → HF→ dementia: HR = 1.48, 95% CI = 1.03–2.13), and the transition from HF to dementia (pathway baseline→ MI → HF→ dementia: HR = 1.59, 95% CI = 1.22–2.07) (Tables 2–5). In the sensitivity analyses, the results were not substantially altered after excluding CVD and dementia during the first two years of follow-up (Supplementary Tables 20–23).
Table 3.
The associations between components of Life’s Crucial 9 score and dynamic transitions of myocardial infarction, heart failure, and dementia
| Transition | Behavior scale | Biological scale | Psychological health disorder | |
|---|---|---|---|---|
| No | Yes | |||
| State 1 to State 2 | ||||
| Baseline → Myocardial infarction | 0.78 (0.77–0.80) | 0.68 (0.66–0.69) | 1.00 (REF) | 1.12 (1.06–1.17) |
| State 1 to State 3 | ||||
| Baseline → Heart failure | 0.70 (0.69–0.72) | 0.79 (0.77–0.81) | 1.00 (REF) | 1.35 (1.28–1.43) |
| State 1 to State 4 | ||||
| Baseline → Dementia | 1.00 (0.97–1.03) | 0.96 (0.93–0.99) | 1.00 (REF) | 1.53 (1.44–1.64) |
| State 2 to State 3 | ||||
| Myocardial infarction → Heart failure | 0.95 (0.90–1.00) | 0.88 (0.83–0.93) | 1.00 (REF) | 1.27 (1.11–1.44) |
| State 2 to State 4 | ||||
| Myocardial infarction → Dementia | 0.86 (0.74–1.01) | 0.85 (0.71–1.01) | 1.00 (REF) | 1.48 (1.03–2.13) |
| State 3 to State 4 | ||||
| Heart failure → Dementia | 0.97 (0.86–1.08) | 1.02 (0.89–1.16) | 1.00 (REF) | 1.59 (1.22–2.07) |
The model adjusted for age, sex, ethnicity, education, Townsend deprivation index, and current drinking. The hazard ratios (95% confidence intervals) for the associations between components of Life’s Crucial 9 and specific transition are given in the table. Statistically significant results are bolded.
REF reference. Compared with participants without psychological health disorder.
Table 4.
The associations between components of Life’s Crucial 9 score and dynamic transitions of atrial fibrillation, heart failure, and dementia
| Transition | Behavior scale | Biological scale | Psychological health disorder | |
|---|---|---|---|---|
| No | Yes | |||
| State 1 to State 2 | ||||
| Baseline → Atrial fibrillation | 0.85 (0.84–0.87) | 0.92 (0.91–0.94) | 1.00 (REF) | 1.06 (1.02–1.10) |
| State 1 to State 3 | ||||
| Baseline → Heart failure | 0.68 (0.66–0.70) | 0.72 (0.70–0.74) | 1.00 (REF) | 1.50 (1.41–1.61) |
| State 1 to State 4 | ||||
| Baseline → Dementia | 0.99 (0.96–1.02) | 0.97 (0.94–1.00) | 1.00 (REF) | 1.59 (1.48–1.70) |
| State 2 to State 3 | ||||
| Atrial fibrillation → Heart failure | 0.83 (0.80–0.87) | 0.88 (0.84–0.93) | 1.00 (REF) | 1.31 (1.18–1.46) |
| State 2 to State 4 | ||||
| Atrial fibrillation → Dementia | 0.89 (0.81–0.96) | 0.85 (0.77–0.93) | 1.00 (REF) | 1.45 (1.17–1.79) |
| State 3 to State 4 | ||||
| Heart failure → Dementia | 0.99 (0.87–1.12) | 1.06 (0.92–1.22) | 1.00 (REF) | 1.54 (1.16–2.05) |
The model adjusted for age, sex, ethnicity, education, Townsend deprivation index, and current drinking. The hazard ratios (95% confidence intervals) for the associations between components of Life’s Crucial 9 and specific transition are given in the table. Statistically significant results are bolded.
REF reference. Compared with participants without psychological health disorder.
Table 2.
The associations between components of Life’s Crucial 9 score and dynamic transitions of atrial fibrillation, ischemic stroke, and dementia
| Transition | Behavior scale | Biological scale | Psychological health disorder | |
|---|---|---|---|---|
| No | Yes | |||
| State 1 to State 2 | ||||
| Baseline → Atrial fibrillation | 0.85 (0.83–0.86) | 0.92 (0.90–0.93) | 1.00 (REF) | 1.08 (1.04–1.12) |
| State 1 to State 3 | ||||
| Baseline → Ischemic stroke | 0.80 (0.78–0.82) | 0.76 (0.73–0.78) | 1.00 (REF) | 1.27 (1.18–1.36) |
| State 1 to State 4 | ||||
| Baseline → Dementia | 0.99 (0.96–1.02) | 0.98 (0.95–1.01) | 1.00 (REF) | 1.61 (1.51–1.73) |
| State 2 to State 3 | ||||
| Atrial fibrillation → Ischemic stroke | 0.97 (0.91–1.03) | 0.85 (0.79–0.91) | 1.00 (REF) | 0.90 (0.76–1.07) |
| State 2 to State 4 | ||||
| Atrial fibrillation → Dementia | 0.94 (0.86–1.02) | 0.89 (0.81–0.98) | 1.00 (REF) | 1.48 (1.20–1.81) |
| State 3 to State 4 | ||||
| Ischemic stroke → Dementia | 0.90 (0.79–1.02) | 0.80 (0.69–0.93) | 1.00 (REF) | 1.39 (1.02–1.89) |
The model adjusted for age, sex, ethnicity, education, Townsend deprivation index, and current drinking. The hazard ratios (95% confidence intervals) for the associations between components of Life’s Crucial 9 and specific transition are given in the table. Statistically significant results are bolded.
REF reference. Compared with participants without psychological health disorder.
Table 5.
The associations between components of Life’s Crucial 9 score and dynamic transitions of atrial fibrillation, myocardial infarction, and dementia
| Transition | Behavior scale | Biological scale | Psychological health disorder | |
|---|---|---|---|---|
| No | Yes | |||
| State 1 to State 2 | ||||
| Baseline → Atrial fibrillation | 0.86 (0.85–0.87) | 0.93 (0.91–0.94) | 1.00 (REF) | 1.07 (1.03–1.11) |
| State 1 to State 3 | ||||
| Baseline → Myocardial infarction | 0.76 (0.75–0.78) | 0.67 (0.66–0.69) | 1.00 (REF) | 1.15 (1.10–1.21) |
| State 1 to State 4 | ||||
| Baseline → Dementia | 1.00 (0.97–1.03) | 0.97 (0.94–1.01) | 1.00 (REF) | 1.62 (1.51–1.73) |
| State 2 to State 3 | ||||
| Atrial fibrillation → Myocardial infarction | 0.84 (0.80–0.89) | 0.72 (0.68–0.76) | 1.00 (REF) | 1.33 (1.17–1.51) |
| State 2 to State 4 | ||||
| Atrial fibrillation → Dementia | 0.93 (0.86–1.01) | 0.86 (0.78–0.95) | 1.00 (REF) | 1.47 (1.20–1.80) |
| State 3 to State 4 | ||||
| Myocardial infarction → Dementia | 0.86 (0.76–0.97) | 0.87 (0.75–1.00) | 1.00 (REF) | 1.43 (1.06–1.94) |
The model adjusted for age, sex, ethnicity, education, Townsend deprivation index, and current drinking. The hazard ratios (95% confidence intervals) for the associations between components of Life’s Crucial 9 and specific transition are given in the table. Statistically significant results are bolded.
REF reference. Compared with participants without psychological health disorder.
Association between LC9 and incident dementia among CVD-prevalent participants
Among 18,470 CVD-prevalent participants, 925 incident dementia cases were recorded during a median follow-up of 13 years. The risk of dementia among CVD patients with higher LC9 scores was significantly lower than in those with low LC9 scores, especially in IS patients (HR for high LC9 scores=0.15, 95% CI = 0.03–0.66) (Supplementary Table 24). In addition, per SD increment in behavior and biological subscale scores was associated with risk reductions of 12% to 25% in incident dementia, except for patients with AF. Similarly, psychological health disorders also played a crucial role in the development of dementia among CVD patients, and the strongest association was observed in HF patients (HR = 1.75, 95% CI = 1.15–2.66).
Discussion
By using the UK Biobank data, we found that LC9 played an important role in the development of CVD or dementia, and dynamic transitions of CVD and dementia, but to different extents. Of note, the biopsychosocial subscale contributed more strongly to all dynamic transitions from CVD to dementia than the behavior subscale. In addition, CVD patients with higher LC9 scores still had a significantly lower risk of dementia in the CVD-prevalent cohort.
Although the AHA writing group acknowledged the role of psychological health in promoting cardiovascular health, there is still a lack of agreement on how to combine psychological health factors. LC9 was just a proposal in a recent perspective11. To our knowledge, this is the first study to explore the associations of the new cardiovascular health metrics, defined by LC9, with the risk of CVD, dementia, and dynamic transitions of CVD and dementia (i.e., the transitions from free of CVD to AF incidence, subsequently to IS, HF, or MI, and further to dementia; the transitions from free of CVD to incident MI, HF, and dementia). Overall, participants with higher LC9 scores had a lower risk of developing CVD or dementia. In addition, the behavior and biological subscale were more specifically related to the risk of CVD, such as MI and HF. In contrast, we observed that psychological health disorders, including depression, anxiety, chronic and traumatic stress, and social integration, mainly contributed to dementia incidence. Consistent with previous findings, metabolic factors were the predominant risk factors for CVD21. Similarly, a meta-analysis study also suggested that social isolation might be a more important risk factor for dementia, compared to hypertension, diabetes, and physical inactivity22.
Accumulating epidemiological studies have found that CVD often coexists with dementia in the aging population, which could probably be explained by cerebral hypoperfusion and hypoxia, thromboembolic complications, direct effects of natriuretic peptides, or amyloid-beta (Aβ) peptides23. Therefore, in our study, we used the multi-state model to explore the role of LC9 on dynamic trajectories from baseline to CVD and progression to dementia, and identify which component of LC9 might mainly contribute to different dynamic transitions. We found that combined LC9 scores were associated with all possible transitions, except for the transitions from MI or HF to dementia, suggesting that LC9 also plays a crucial role in the comorbidity of CVD and dementia. Previous studies also indicated that several vascular risk factors were associated with an increased risk of dementia13,24, and control of these factors might effectively reduce the risk of dementia incidence25. One of the most important components of the Alzheimer’s disease amyloid hypothesis is tissue inflammation and organ dysfunction due to the production and deposition of Αβ peptides, while adherence to healthy lifestyles might alter Αβ peptides metabolism26. As for the negative findings in the transitions from MI or HF to dementia, this might be partly explained by poorer survival in patients with both MI and HF, which might preclude the excess risk of dementia incidence27. Nevertheless, our findings imply the significance of strengthening comprehensive interventions in the primary and secondary prevention of CVD and dementia.
Of note, we further found that compared to the behavior and biological subscale, the psychological subscale might contribute more strongly to all dynamic transitions from CVD to dementia, especially in the transitions from MI or HF to dementia. Recently, the effects of psychological health on the risk of CVD and dementia have received increasing attention. Some evidence suggested that individuals with depression had a higher risk of HF, while HF also increased the risk of depression28. Moreover, HF and depression were also independently associated with cognitive impairment. Psychological health disorders, such as major depressive disorder (MDD), post-traumatic stress disorder, and anxiety, might be associated with an increase in Aβ peptides deposition29,30. Compared to those with dementia alone, patients with both dementia and MDD have increased rates of Aβ peptides production and deposition. In addition, several studies also indicated that better perceived social support was associated with better brain structure31, while loneliness was associated with smaller hippocampus, gray matter volumes of the amygdala, and entorhinal cortex32.
In the CVD-prevalent cohort, the overall lower risk of dementia was still observed among CVD patients with higher LC9 scores, suggesting that better LC9-defined cardiovascular health plays an important role in the prevention of dementia regardless of patients with or without a history of CVD. Moreover, the patients with CVD diagnosis before cohort entry were more likely to have psychological health disorders in our study. Similarly, psychological health disorders were also associated with a higher risk of dementia among CVD patients, especially in HF patients. Previous studies have also shown that comorbid depression and CVD were associated with poor prognosis and worse quality of life in CVD patients28,33. However, pharmacologic and psychotherapeutic treatment might mitigate the risk of major cardiovascular adverse events. Despite ACC/AHA guideline recommending routine screening for mental health in patients with chronic coronary disease, treatment with antidepressants and psychotherapies is still low in patients after cardiovascular events34. With the improving life expectancy of CVD patients, psychological health disorders have been non-negligible contributors to the development of dementia following a CVD event, and the need for its surveillance is significant to reduce the increasing medication cost and burden of care.
The major strength of our study is the use of multi-state model to investigate the roles of the new cardiovascular health metrics on different trajectories of progression from baseline to CVD and dementia. Furthermore, the large sample size of the UK Biobank provided substantial power to explore these dynamic transitions. Some limitations should also be noted. First, the information on behavioral factors was mainly obtained by self-report, and recall bias was inevitable. However, we would assume the error in the recall of this information might bias the findings to the null, and yet significant associations were still detected in this study. Second, LC9 scores were calculated based on baseline information, and we did not account for potential changes during follow-up. Therefore, the exposure measurement misclassifications could not be fully excluded. However, using baseline information to construct LC9 scores might help avoid reverse causation due to lifestyle change after disease onset. In addition, the results remained consistent after restricting analysis among participants with a CVD diagnosis before cohort entry since we assumed that the lifestyle might change drastically due to the awareness of CVD. Third, LC9 integrates psychological health into cardiovascular health metrics as a new component. However, there is no specific scoring criteria for LC9, the categories of LC9 based on the AHA advisory for LE8 might not be fully applicable, which might introduce measurement bias. In our study, LC9 scores were analyzed as both categorical variable and standardized continuous variable, and the results consistently showed that participants with higher LC9 scores had a lower risk of developing CVD or dementia. Fourth, information on the treatment for CVD and dementia was not available in the UK Biobank, medication compliance might be associated with adherence to a healthy lifestyle. However, we have adjusted education and Townsend deprivation index as proxies for several health-related behaviors. Fifth, the disease prevalence and incidence rates, and rates of disease transition were relatively lower in the UK Biobank due to the healthy volunteer bias, for example, 930 (5.1%) AF patients further developed IS. Compared with the general population, participants in the UK Biobank were more likely to live in less socioeconomically deprived areas, less likely to be obese, and to smoke35. Therefore, our findings need to be validated by further studies in the real world.
Conclusion
In conclusion, by using data from a large cohort, we found that better LC9-defined cardiovascular health was associated with lower risks of dynamic transitions from baseline to CVD and dementia, suggesting the importance of comprehensive interventions in the primary and secondary prevention of CVD and dementia. In addition, the biopsychosocial subscale, especially psychological health, might contribute more strongly to all dynamic transitions from CVD to dementia, further highlighting the significance of psychological health surveillance in CVD patients to lower the risk of subsequent dementia incidence.
Supplementary information
Description of Additional Supplementary files
Acknowledgements
We appreciate efforts made by the original data creators, depositors, copyright holders, the funders of the data collections, and their contributions to access to data from the UK Biobank, approved project number 90492. This work is supported by the National Natural Science Foundation of China (82373665), the Capital’s Funds for Health Improvement and Research (CFH 2024-2G-4253), the Non-profit Central Research Institute Fund of Chinese Academy of Medical Sciences (2021-RC330-001), and the 2022 China Medical Board-Open Competition research grant (22-466). The sponsor of the study had no role in the study design, data collection, data analysis, data interpretation, or writing of the report.
Author contributions
W.X. and F.Z. had full access to the data, and took responsibility for the integrity of the data and the accuracy of the analysis. Study concept and design: W.X., Y.Z. Acquisition, analysis, and interpretation of data: Y.Z., W.X. Drafting of the manuscript: Y.Z. Critical revision of the manuscript for important intellectual content: All authors. Statistical analysis: Y.Z. Funding: W.X. and F.Z.
Peer review
Peer review information
Communications Medicine thanks Silvan Licher and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Data availability
This study was conducted using the UK Biobank Resource under Application 90492. Wuxiang Xie and Fanfan Zheng have full access to the data in the study. The UK Biobank data are protected and are not available due to data privacy laws. Researchers interested in accessing the data used in this study can apply for access to the UK Biobank by visiting website (https://www.ukbiobank.ac.uk/enable-your-research/apply-for-access) and submitting an application that includes a research protocol summary and requested data fields. Upon approval by the UK Biobank management team and payment of applicable fees, researchers will be granted access to the database. Source data are provided with this paper. The source data for main text Fig. 1 is included in Supplementary Data 3. The source data for main text Fig. 2 is included in Supplementary Tables 6–9 and Supplementary Tables 12–15. The source data for Supplementary Figs. 2 and 3 are included in Supplementary Data 4 and 5, respectively. STROBE statement is presented in Supplementary Data 6.
Code availability
Multi-state model was conducted using the Stata package of multistate, and commands were available at this article (10.1002/sim.7448)36. Restricted cubic splines were developed using the R package of rms (https://cran.r-project.org/web/packages/rms), and the Fine and Gray models were conducted using the R package of cmprsk (https://cran.r-project.org/web/packages/cmprsk). All other codes for the analyses will be made available upon request.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Contributor Information
Fanfan Zheng, Email: zhengfanfan@nursing.pumc.edu.cn.
Wuxiang Xie, Email: xiewuxiang@hsc.pku.edu.cn.
Supplementary information
The online version contains supplementary material available at 10.1038/s43856-025-00938-z.
References
- 1.Global burden of 288 causes of death and life expectancy decomposition in 204 countries and territories and 811 subnational locations, 1990–2021: a systematic analysis for the Global Burden of Disease Study 2021. Lancet10.1016/s0140-6736(24)00367-2 (2024). [DOI] [PMC free article] [PubMed]
- 2.Zhang, W. et al. Age at diagnosis of atrial fibrillation and incident dementia. JAMA Netw. Open6, e2342744 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Livingston, G. et al. Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet396, 413–446 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Tsao, C. W. et al. Heart disease and stroke statistics-2022 update: a report from the American Heart Association. Circulation145, e153–e639 (2022). [DOI] [PubMed] [Google Scholar]
- 5.Bunch, T. J. Atrial fibrillation and dementia. Circulation142, 618–620 (2020). [DOI] [PubMed] [Google Scholar]
- 6.Hall, M. et al. Health outcomes after myocardial infarction: a population study of 56 million people in England. PLoS Med.21, e1004343 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Vishwanath, S. et al. Cognitive trajectories and incident dementia after a cardiovascular event in older adults. Alzheimers Dement19, 3670–3678 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Zheng, F. et al. Age at onset of heart failure and subsequent risk of dementia. A longitudinal cohort study. JACC Heart Fail.12, 826–835 (2024). [DOI] [PubMed] [Google Scholar]
- 9.Levine, G. N. et al. Psychological health, well-being, and the mind-heart-body connection: a scientific statement from the American Heart Association. Circulation143, e763–e783 (2021). [DOI] [PubMed] [Google Scholar]
- 10.Yang, L. et al. Depression, depression treatments, and risk of incident dementia: a prospective cohort study of 354,313 participants. Biol. Psychiatry93, 802–809 (2023). [DOI] [PubMed] [Google Scholar]
- 11.Gaffey, A. E., Rollman, B. L. & Burg, M. M. Strengthening the pillars of cardiovascular health: psychological health is a crucial component. Circulation149, 641–643 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Lloyd-Jones, D. M. et al. Life’s Essential 8: updating and enhancing the American Heart Association’s construct of cardiovascular health: a presidential advisory from the American Heart Association. Circulation146, e18–e43 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Qiu, C. & Fratiglioni, L. A major role for cardiovascular burden in age-related cognitive decline. Nat. Rev. Cardiol.12, 267–277 (2015). [DOI] [PubMed] [Google Scholar]
- 14.Han, Y. et al. Lifestyle, cardiometabolic disease, and multimorbidity in a prospective Chinese study. Eur. Heart J.42, 3374–3384 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Luo, H. et al. Long-term exposure to ambient air pollution is a risk factor for trajectory of cardiometabolic multimorbidity: a prospective study in the UK Biobank. EBioMedicine84, 104282 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Li, X., Ma, H., Wang, X., Feng, H. & Qi, L. Life’s Essential 8, genetic susceptibility, and incident cardiovascular disease: a prospective study. Arterioscler. Thromb. Vasc. Biol.43, 1324–1333 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Tang, R. et al. Adherence to Life’s Essential 8 and incident chronic kidney disease: a prospective study of 147,988 UK Biobank participants. Am. J. Clin. Nutr.118, 804–811 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Bradbury, K. E., Young, H. J., Guo, W. & Key, T. J. Dietary assessment in UK Biobank: an evaluation of the performance of the touchscreen dietary questionnaire. J. Nutr. Sci.7, e6 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Petermann-Rocha, F. et al. Nonlinear associations between cumulative dietary risk factors and cardiovascular diseases, cancer, and all-cause mortality: a prospective cohort study from UK biobank. Mayo Clin. Proc.96, 2418–2431 (2021). [DOI] [PubMed] [Google Scholar]
- 20.de Wreede, L. C., Fiocco, M. & Putter, H. The mstate package for estimation and prediction in non- and semi-parametric multi-state and competing risks models. Comput Methods Prog. Biomed.99, 261–274 (2010). [DOI] [PubMed] [Google Scholar]
- 21.Yusuf, S. et al. Modifiable risk factors, cardiovascular disease, and mortality in 155 722 individuals from 21 high-income, middle-income, and low-income countries (PURE): a prospective cohort study. Lancet395, 795–808 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Desai, R., John, A., Stott, J. & Charlesworth, G. Living alone and risk of dementia: a systematic review and meta-analysis. Ageing Res. Rev.62, 101122 (2020). [DOI] [PubMed] [Google Scholar]
- 23.Wolters, F. J. et al. Coronary heart disease, heart failure, and the risk of dementia: a systematic review and meta-analysis. Alzheimers Dement.14, 1493–1504 (2018). [DOI] [PubMed] [Google Scholar]
- 24.Gottesman, R. F. et al. Associations between midlife vascular risk factors and 25-year incident dementia in the Atherosclerosis Risk in Communities (ARIC) cohort. JAMA Neurol.74, 1246–1254 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.de la Torre, J. C. Vascular risk factor detection and control may prevent Alzheimer’s disease. Ageing Res. Rev.9, 218–225 (2010). [DOI] [PubMed] [Google Scholar]
- 26.Stakos, D. A. et al. The Alzheimer’s disease amyloid-beta hypothesis in cardiovascular aging and disease: JACC focus seminar. J. Am. Coll. Cardiol.75, 952–967 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Imahori, Y. et al. Association of ischemic heart disease with long-term risk of cognitive decline and dementia: a cohort study. Alzheimers Dement.19, 5541–5549 (2023). [DOI] [PubMed] [Google Scholar]
- 28.Sbolli, M., Fiuzat, M., Cani, D. & O’Connor, C. M. Depression and heart failure: the lonely comorbidity. Eur. J. Heart Fail.22, 2007–2017 (2020). [DOI] [PubMed] [Google Scholar]
- 29.Pink, A. et al. A longitudinal investigation of Aβ, anxiety, depression, and mild cognitive impairment. Alzheimers Dement.18, 1824–1831 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Rafferty, L. A., Cawkill, P. E., Stevelink, S. A. M., Greenberg, K. & Greenberg, N. Dementia, post-traumatic stress disorder and major depressive disorder: a review of the mental health risk factors for dementia in the military veteran population. Psychol. Med48, 1400–1409 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.van der Velpen, I. F. et al. Social health is associated with structural brain changes in older adults: the Rotterdam study. Biol. Psychiatry Cogn. Neurosci. Neuroimaging7, 659–668 (2022). [DOI] [PubMed] [Google Scholar]
- 32.Düzel, S. et al. Structural brain correlates of loneliness among older adults. Sci. Rep.9, 13569 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Hare, D. L., Toukhsati, S. R., Johansson, P. & Jaarsma, T. Depression and cardiovascular disease: a clinical review. Eur. Heart J.35, 1365–1372 (2014). [DOI] [PubMed] [Google Scholar]
- 34.Virani, S. S. et al. 2023 AHA/ACC/ACCP/ASPC/NLA/PCNA guideline for the management of patients with chronic coronary disease: a report of the American Heart Association/American College of Cardiology Joint Committee on clinical practice guidelines. Circulation148, e9–e119 (2023). [DOI] [PubMed] [Google Scholar]
- 35.Fry, A. et al. Comparison of sociodemographic and health-related characteristics of UK biobank participants with those of the general population. Am. J. Epidemiol.186, 1026–1034 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Crowther, M. J. & Lambert, P. C. Parametric multistate survival models: flexible modelling allowing transition-specific distributions with application to estimating clinically useful measures of effect differences. Stat. Med.36, 4719–4742 (2017). [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Description of Additional Supplementary files
Data Availability Statement
This study was conducted using the UK Biobank Resource under Application 90492. Wuxiang Xie and Fanfan Zheng have full access to the data in the study. The UK Biobank data are protected and are not available due to data privacy laws. Researchers interested in accessing the data used in this study can apply for access to the UK Biobank by visiting website (https://www.ukbiobank.ac.uk/enable-your-research/apply-for-access) and submitting an application that includes a research protocol summary and requested data fields. Upon approval by the UK Biobank management team and payment of applicable fees, researchers will be granted access to the database. Source data are provided with this paper. The source data for main text Fig. 1 is included in Supplementary Data 3. The source data for main text Fig. 2 is included in Supplementary Tables 6–9 and Supplementary Tables 12–15. The source data for Supplementary Figs. 2 and 3 are included in Supplementary Data 4 and 5, respectively. STROBE statement is presented in Supplementary Data 6.
Multi-state model was conducted using the Stata package of multistate, and commands were available at this article (10.1002/sim.7448)36. Restricted cubic splines were developed using the R package of rms (https://cran.r-project.org/web/packages/rms), and the Fine and Gray models were conducted using the R package of cmprsk (https://cran.r-project.org/web/packages/cmprsk). All other codes for the analyses will be made available upon request.