Association between ADHD history and “Life’s Crucial 9” cardiovascular risk factors in young adults Authors: Na Li, Meiyuan Chen, Sidong Li, Yang Du, Yuxiao Wu, Xiaodong Zang, Ruishu Zhang, Zihan Zhang, Buyun Liu, Wei Bao, Guifeng Xu Published: 2026 ABSTRACT ObjectiveSymptoms of attention-deficit/hyperactivity disorder (ADHD) often persist into adulthood, increasing cardiovascular disease (CVD) risk. This study extends our previous research, by examining associations between ADHD and modifiable CVD risk factors, incorporating the recently proposed Life's Crucial 9 (LC9) metric.MethodsWe conducted a prospective analysis using data from the National Longitudinal Study of Adolescent to Adult Health (Add Health). ADHD was identified at Wave IV (mean age 28.4), and at Wave V (approximately 9 years later), we assessed nine CVD risk components, including smoking, physical activity, diet, sleep, body mass index, lipids, blood glucose, blood pressure, and psychological health status A cumulative LC9 risk factor count (range 0–9) was calculated by summing adverse components, with higher counts indicating greater LC9 risk factor burden. Survey-weighted linear, logistic, and multinomial logistic regression models were used to estimate these associations, adjusting for potential confounders.ResultsAmong 10 913 participants (mean age 28.4 years; 49.4 % male), 492 (4.7 %) reported an ADHD diagnosis. Over 9 years of follow-up, ADHD was associated with a higher LC9 risk factor count. Compared with individuals without ADHD, those with ADHD had higher odds of cigarette smoking (OR, 1.66 [95 % CI, 1.26–2.18]), sleep problems (OR, 1.41 [95 % CI, 1.09–1.82]), and anxiety and/or depression (OR, 3.31 [95 % CI, 2.50–4.42]). In joint-category analyses, participants with both ADHD and anxiety and/or depression showed the most adverse profiles, particularly for smoking, sleep problems, hyperlipidemia, and hypertension.ConclusionsADHD in young adulthood is associated with an unfavorable cardiovascular risk profile. Psychological comorbidity substantially amplifies this burden, highlighting the importance of integrating mental health screening into LC9-based cardiovascular prevention strategies. Central illustration.Unlabelled image dummy alt text Central illustration. Introduction Attention deficit/hyperactivity disorder (ADHD) is a prevalent neurodevelopmental disorder, affecting approximately 10.2 % of U.S. children and adolescents and around 5 % of children globally [1,2]. Previous studies, including our own, have found a significant association between a history of ADHD diagnosis and cardiovascular disease (CVD) in adulthood [3], identifying ADHD as an independent risk factor for CVD in adulthood [4]. However, research characterizing comprehensive cardiovascular risk profiles in young adults with ADHD remains scarce. CVD remains the leading cause of morbidity and mortality in the US, accounting for over one in four deaths [5]. To promote cardiovascular health, the American Heart Association (AHA) introduced "Life’s Simple 7," which was later expanded to "Life’s Essential 8" with the inclusion of sleep health [6,7]. More recently, scholars have further refined the framework into "Life’s Crucial 9" (LC9) by incorporating psychological health as an additional key component [8]. Psychological disorders are highly prevalent in the general population, and their global burden continues to rise [9]. Accumulating evidence supports that psychological disorders are significantly associated with CVD and are linked to higher CVD mortality and poorer prognosis, with the risk of adverse outcomes being approximately 50 % to 2-fold higher than among those without psychological disorders [[9], [10], [11]]. Recent studies have shown that adding psychological health to the Life’s Essential 8 framework to form Life’s Crucial 9 (LC9) provides only modest improvement in predicting cardiovascular and all-cause mortality, suggesting that the impact of psychological health may differ across populations [12,13]. Numerous studies have indicated that psychological disorders are commonly comorbid with ADHD [14]. Therefore, when evaluating the increased risk of CVD in individuals with ADHD, it is essential to consider common psychological comorbidities, such as anxiety and depression. Previous studies have examined associations between ADHD and individual or multiple CVD risk factors [15]. For instance, a prospective UK birth cohort study showed that ADHD was associated with higher body mass index, blood pressure and triglyceride levels, as well as higher odds of being a current smoker [16]. Our team’s prior studies identified significant associations of ADHD diagnosis with tobacco use (including e-cigarette use) and diabetes [17,18].Furthermore, additional studies indicated that individuals with ADHD have a significantly higher risk of having sleep disturbances, consuming foods with lower nutritional density and higher total fat content, and engaging in less physical activity compared to their non-ADHD individuals [19]. However, to date, no study has directly evaluated the association between ADHD and the comprehensive LC9 framework. Furthermore, research on the impact of psychological health distress on CVD risk among individuals with ADHD remains markedly lacking. To address these critical gaps, we utilized a nationally representative cohort to examine the association between ADHD and both the overall LC9 risk factor burden (defined as the count of adverse LC9 components) and each individual cardiovascular risk factor. Methods Study population The National Longitudinal Study of Adolescent to Adult Health (Add Health), initiated in 1994, enrolled a nationally representative sample of U.S. adolescents in grades 7–12 [20]. The cohort has been followed into adulthood through six waves of data collection (1994–95, 1996, 2001–02, 2008, 2016–18, and 2022–25).The study was supported by NICHD and co-funded by 23 federal agencies and foundations. Add Health combines longitudinal survey data collected from school-based survey and in-home interviews on respondents’ social, economic, psychological and physical well-being with contextual data on the family, neighborhood, community, school, friendships, peer groups, and romantic relationships, as well as biological data. In the prospective analysis, 10 913 participants at Wave IV were included and followed through Wave V. Race/ethnicity was self-reported at Wave I and treated as a baseline covariate. Ascertainment of ADHD ADHD diagnosis was determined based on self-report in Wave IV, where participants were asked: “Has a doctor, nurse, or other healthcare provider ever told you that you have or had attention problems or ADD or ADHD?”. Respondents who answered “Yes” were classified as having ADHD, while those who answered “No” were classified as not having ADHD. Ascertainment of “Life’s Crucial 9” risk factors for CVD Smoking status was classified into three categories based on self-reported data. Never smokers reported no tobacco use history; former smokers reported prior smoking but no cigarette or e-cigarette use in the past 30 days; current smokers reported cigarette or e-cigarette use within the past 30 days. Physical activity was assessed by self-reported participation in moderate-to-vigorous physical activity (MVPA) during the past week and whether this represented their typical physical activity level. MVPA was categorized as none, insufficient (<5 bouts/week), or sufficient (≥5 bouts/week) [15]. Participants who reported no MVPA were classified as having low levels of physical activity. Information about dietary habit included consumption of fast food and sweetened drinks. We categorized the dietary habit into four groups: fast food < 3/week and sweetened drinks < 2/day, fast food < 3/week and sweetened drinks ≥ 2/day, fast food ≥ 3/week and sweetened drinks < 2/day, fast food ≥3/week and sweetened ≥ 2/day (defined as exhibiting unhealthy dietary habits due to the combination of frequent fast-food consumption and high intake of sweetened beverages). Body mass index (BMI) was calculated from anthropometric measurements in Wave V. Overweight was defined as BMI 25–29.9 kg/m² and obesity as BMI ≥ 30 kg/m². Average sleep duration was calculated from self-reported data and categorized based on established sleep guidelines recommending approximately 7–9 h per night for adults and young adults: normal (7–9 h), insufficient (<7 h), or excessive (>9 h) [21]. Diabetes was defined according to a self-reported history of diagnosis, anti-diabetic medication use, fasting blood glucose levels ≥ 126 mg/dl or non-fasting blood glucose levels ≥ 200 mg/dl, or HbA1c levels ≥ 6.5 %. Hyperlipidemia was defined based on self-reported physician diagnosis, use of antihyperlipidemic medications, or clinical criteria of fasting triglycerides (TG) ≥500 mg/dL or low-density lipoprotein cholesterol (LDL-C) ≥190 mg/dL. Hypertension was defined as a self-reported physician diagnosis, current use of antihypertensive medications within the past 4 weeks, or clinical measurements of systolic blood pressure ≥140 mmHg or diastolic blood pressure ≥90 mmHg. Psychological distress was defined as self-reported physician diagnoses of depression or anxiety, with the use of related medications also considered at Wave V. Further details on questionnaire items and variable definitions from the Add Health study are provided in Table S1. Operationalization of LC9 risk factors In this study, the nine LC9 domains were operationalized as dichotomized cardiovascular and psychological risk factors rather than a continuous AHA cardiovascular health score. Each domain was coded as 0 (absence of an adverse profile) or 1 (presence of an adverse profile) based on predefined thresholds adapted from Add Health measures. This binary approach was adopted because several domains in Add Health, including diet, smoking, physical activity, blood glucose, lipids, and psychological health, were not recorded with the granularity required to derive official, non-binary AHA-style scores. Although some domains (e.g., blood pressure and sleep duration) could be represented on a more graded scale, applying different scoring approaches across domains would compromise comparability and bias the summary measure. We therefore used a consistent binary framework across all nine domains. The individual scores were summed to generate an LC9 risk factor count (range 0–9), where higher values indicate a greater burden of adverse cardiovascular profiles. For descriptive purposes, this count was further grouped as 0–1, 2, 3, and ≥4 according to the observed distribution. This categorization was intended to facilitate the interpretation of cumulative risk burden rather than to define validated clinical thresholds. Accordingly, throughout the manuscript, this measure is referred to as the “LC9 risk factor count” or “LC9 risk factor burden” Assessment of covariates Information about age, sex, race/ethnicity, education, family income, and alcohol drinking were collected through questionnaires. Sex was classified as male or female according to self-reported biological sex. Race and ethnicity were classified based on the 1997 standards by the US Office of Management and Budget. Education attainment was categorized into less than high school, High school or General Educational Development (GED) or Vocational training, and college or above. Family income was classified into four levels: <25,000, 25,000–49,999, 50,000–74,999, and ≥75,000 USD annually. Alcohol consumption was categorized based on responses to two self-reported questions: "Have you ever had a drink of beer, wine, or liquor more than two or three times (excluding sips or tastes from someone else's drink)?" and "During the past 12 months, on how many days did you drink alcohol?" Participants were classified as never drinkers if they reported no lifetime alcohol use; former drinkers if they reported previous drinking but no alcohol consumption in the past year; occasional drinkers if they consumed alcohol two to three times per month or less; and frequent drinkers if they consumed alcohol at least once per week. Statistical analysis Baseline characteristics were compared between participants with and without a history of ADHD using t tests for continuous variables and chi-square tests or Fisher’s exact tests, as appropriate, for categorical variables. Primary analyses were prospective and evaluated whether ADHD status at Wave IV was associated with LC9 risk factors measured at Wave V after an average follow-up of 9 years. Survey-weighted logistic regression was employed for our primary analyses, as the outcome was defined as the presence of binary LC9 risk factors at a single follow-up time point (Wave V). Because our primary aim was to examine whether ADHD at Wave IV was associated with the LC9 risk factor profile at Wave V, rather than to model within-person change over time, baseline levels of the corresponding outcomes were not included in the primary models. The LC9 risk factor count (range, 0–9) was analyzed using survey-weighted linear regression, whereas associations with individual LC9 risk factors were estimated using survey-weighted logistic regression and are reported as odds ratios (ORs) with 95 % confidence intervals (CIs). For descriptive analyses of cumulative risk burden, the LC9 risk factor count was additionally categorized as 0–1, 2, 3 and ≥4 and analyzed using survey-weighted multinomial logistic regression. Covariates were introduced sequentially across models: Model 1 was unadjusted; Model 2 adjusted for sociodemographic characteristics and alcohol consumption; and Model 3 further adjusted for psychological health, except when psychological health was modeled as the outcome or as an effect modifier. Supplemental cross-sectional analyses at Wave IV were conducted for descriptive comparison. To examine the joint associations of ADHD and psychological distress with LC9 risk factors, participants were classified into four groups: non-ADHD without anxiety and/or depression, non-ADHD with anxiety and/or depression, ADHD without anxiety and/or depression, and ADHD with anxiety and/or depression. Interaction terms (ADHD × psychological health) were included in the adjusted models to formally assess potential effect modification. Separately, sex-stratified analyses were performed to estimate sex-specific associations. Additionally, ADHD-by-sex interaction terms were fitted in Models 2 and 3 to formally assess potential sex differences. Sensitivity analyses included additional adjustment for the corresponding baseline outcome level at Wave IV, where comparable measures were available, to assess whether the observed associations were independent of baseline status. In addition, to account for potential overestimation of associations by odds ratios when outcomes are common, we also performed sensitivity analyses using modified Poisson regression to estimate relative risks (RRs) directly and to assess the consistency of the findings across modeling approaches. To ensure that our findings are nationally representative, all analyses were conducted using longitudinal sampling weights provided by Add Health (Wave V weights) to account for differential probabilities of selection and attrition. Analyses were conducted using SAS 9.4 (SAS Institute Inc., Cary, NC, USA) via the SURVEY procedures (e.g., PROC SURVEYLOGISTIC), which incorporate cluster and strata variables. Regarding missing data, in the multivariable models we excluded individuals with missing values for ADHD status, any of the LC9 components, or adjusted key covariates (complete-case analysis approach). All tests were 2-sided, and statistical significance was defined as a two-tailed P value < 0.05. Results Baseline characteristics The Wave IV study population comprised 10,913 young adults (mean age 28.42 [SD 1.78] years; 49.41 % male), among whom 492 (4.72 %) reported a prior diagnosis of ADHD. Compared with participants without ADHD, those with ADHD were more likely to be male, non-Hispanic White, have lower educational attainment, come from lower-income households, and report alcohol use (Table 1).Table 1Baseline characteristics of the study population in the Add Health, according to history of ADHD.Table 1 dummy alt textWave IV (N = 10, 913)Variables/No. of participantsNon-ADHD 10421 (95.49)ADHD 492 (4.72)P valueAge, mean (SD)28.37 (1.78)28.27 (1.76)0.19Gender N (%)Male4330 (49.41)302 (66.80)<.001Female6091 (50.59)190 (33.20)Race/ethnicity N (%)Hispanic1548 (12.55)32 (4.93)<.001Non-Hispanic White6044 (65.49)404 (80.17)Non-Hispanic Black2092 (15.69)35 (6.63)Non-Hispanic Other704 (4.67)18 (3.10)Missing33 (0.35)3 (0.41)Education N (%)Less than high school27 (0.33)7 (1.87)0.005High school or GED or Vocational training3115 (33.39)151 (35.19)College or higher3266 (65.16)327 (61.49)Missing142 (1.12)9 (1.45)Family income N (%)<25,0001459 (15.10)99 (26.12)<.00125,000–49,9991558 (26.98)137 (27.05)50,000–74,9992459 (23.14)101 (16.49)≥75,0003197 (28.44)123 (24.09)Missing592 (6.34)32 (6.24)Alcohol drinking N (%)Non-drinker935 (8.93)51 (14.14)0.012Previous drinker8940 (85.13)405 (78.04)Current drinker494 (5.42)34 (7.34)Missing52 (0.52)2 (0.17)*The P-values represent the statistical comparison between the non-ADHD and ADHD groups for each characteristic.Abbreviation: ADHD, Attention deficit and hyperactivity disorder; SD, Standard Deviation. Table 1Baseline characteristics of the study population in the Add Health, according to history of ADHD.Table 1 dummy alt textWave IV (N = 10, 913)Variables/No. of participantsNon-ADHD 10421 (95.49)ADHD 492 (4.72)P valueAge, mean (SD)28.37 (1.78)28.27 (1.76)0.19Gender N (%)Male4330 (49.41)302 (66.80)<.001Female6091 (50.59)190 (33.20)Race/ethnicity N (%)Hispanic1548 (12.55)32 (4.93)<.001Non-Hispanic White6044 (65.49)404 (80.17)Non-Hispanic Black2092 (15.69)35 (6.63)Non-Hispanic Other704 (4.67)18 (3.10)Missing33 (0.35)3 (0.41)Education N (%)Less than high school27 (0.33)7 (1.87)0.005High school or GED or Vocational training3115 (33.39)151 (35.19)College or higher3266 (65.16)327 (61.49)Missing142 (1.12)9 (1.45)Family income N (%)<25,0001459 (15.10)99 (26.12)<.00125,000–49,9991558 (26.98)137 (27.05)50,000–74,9992459 (23.14)101 (16.49)≥75,0003197 (28.44)123 (24.09)Missing592 (6.34)32 (6.24)Alcohol drinking N (%)Non-drinker935 (8.93)51 (14.14)0.012Previous drinker8940 (85.13)405 (78.04)Current drinker494 (5.42)34 (7.34)Missing52 (0.52)2 (0.17)*The P-values represent the statistical comparison between the non-ADHD and ADHD groups for each characteristic.Abbreviation: ADHD, Attention deficit and hyperactivity disorder; SD, Standard Deviation. Baseline characteristics of the study population in the Add Health, according to history of ADHD. *The P-values represent the statistical comparison between the non-ADHD and ADHD groups for each characteristic. Abbreviation: ADHD, Attention deficit and hyperactivity disorder; SD, Standard Deviation. Association between ADHD and LC9 risk burden At Wave V, participants with ADHD exhibited higher rates of several adverse LC9 components, including current smoking, sleep problems, hypertension, hyperlipidemia, and psychological distress, compared with those without ADHD. ADHD was also associated with a higher overall LC9 risk factor count in both unadjusted and adjusted models (β, 0.58 [95 % CI, 0.34–0.82] in Model 2) (Table 2). In additional descriptive analyses, participants with ADHD showed a less favorable distribution of cumulative LC9 risk burden compared to those without ADHD. In multivariable-adjusted multinomial logistic regression models, ADHD was associated with progressively higher odds of being in the 2, 3, and ≥4 risk factor categories, relative to the 0–1 category (Category 2: OR, 2.80 [95 % CI, 1.23–6.38]; Category 3: OR, 4.09 [95 % CI, 1.96–8.56]; Category≥4: OR, 4.51 [95 % CI, 2.06–9.84]) (Table 3).Table 2Overall LC9 risk burden and distribution of LC9 risk factors at Wave V according to ADHD status.Table 2 dummy alt textProspective (Wave IV–Wave V, N = 10,913)Non-ADHD (10421)ADHD (492)P valueCurrent Smoking N (%)4598 (51.77)320 (70.24)<.001Low physical activity N (%)880 (11.37)50 (12.42)0.66Unhealthy dietary habits N (%)600 (7.74)42 (10.13)0.15Obesity N (%)2198 (47.16)131 (54.58)0.11Sleep problem N (%)4779 (45.68)247 (54.39)0.006Hyperlipidemia N (%)738 (15.80)54 (25.74)0.004Hypertension N (%)1447 (31.52)97 (42.68)0.01Diabetes N (%)427 (9.05)25 (8.02)0.81Psychological health problem N (%)2517 (25.43)246 (49.73)<.001Overall LC9 risk factor burdenMean (95 % CI)2.40 (2.35–2.45)3.23 (2.99–3.47)<0.001Model 1, β coefficient (95 % CI)0.00 (reference)0.83 (0.59–1.08)<0.001Model 2, β coefficient (95 % CI)0.00 (reference)0.58 (0.34–0.82)<0.001aModel 1: Unadjusted Model.bModel 2: Adjusted for age, sex, race, highest education level, annual family income and alcohol consumption.*The P-values represent the statistical comparison between the non-ADHD and ADHD groups for each characteristic.†LC9 risk factor count was calculated as the sum of 9 binary LC9 risk factors, with higher count indicating greater overall risk burden.Abbreviation: ADHD, Attention deficit and hyperactivity disorder.Table 3Cumulative cardiovascular risk burden categories and multinomial logistic regression analyses according to ADHD status at Wave V.Table 3 dummy alt textCumulative LC9 risk factors, N (%)Non-ADHD (N = 3331)ADHD (N = 168)OR (95 % CI)Model 1Model 20–1 risk factors1168 (31.39)26 (12.08)1.00 (reference)1.00 (reference)2 risk factors858 (25.95)38 (23.32)2.34 (1.11–4.93)2.80 (1.23–6.38)3 risk factors617 (19.53)42 (23.70)3.15 (1.55–6.40)4.09 (1.96–8.56)≥4 risk factors688 (23.13)62 (40.90)4.60 (2.31–9.15)4.51 (2.06–9.84)aModel 1: Unadjusted Model.bModel 2: Adjusted for age, sex, race, highest education level, annual family income and alcohol consumption.*The cumulative LC9 risk factor analyses were restricted to participants with complete data on all 9 LC9 components (Non-ADHD: n = 3331; ADHD: n = 168). Percentages are weighted percentages.†Odds ratios were estimated using multinomial logistic regression, with 0–1 LC9 risk factors as the reference category.Abbreviation: ADHD, Attention deficit and hyperactivity disorder; OR, odds ratio. Table 2Overall LC9 risk burden and distribution of LC9 risk factors at Wave V according to ADHD status.Table 2 dummy alt textProspective (Wave IV–Wave V, N = 10,913)Non-ADHD (10421)ADHD (492)P valueCurrent Smoking N (%)4598 (51.77)320 (70.24)<.001Low physical activity N (%)880 (11.37)50 (12.42)0.66Unhealthy dietary habits N (%)600 (7.74)42 (10.13)0.15Obesity N (%)2198 (47.16)131 (54.58)0.11Sleep problem N (%)4779 (45.68)247 (54.39)0.006Hyperlipidemia N (%)738 (15.80)54 (25.74)0.004Hypertension N (%)1447 (31.52)97 (42.68)0.01Diabetes N (%)427 (9.05)25 (8.02)0.81Psychological health problem N (%)2517 (25.43)246 (49.73)<.001Overall LC9 risk factor burdenMean (95 % CI)2.40 (2.35–2.45)3.23 (2.99–3.47)<0.001Model 1, β coefficient (95 % CI)0.00 (reference)0.83 (0.59–1.08)<0.001Model 2, β coefficient (95 % CI)0.00 (reference)0.58 (0.34–0.82)<0.001aModel 1: Unadjusted Model.bModel 2: Adjusted for age, sex, race, highest education level, annual family income and alcohol consumption.*The P-values represent the statistical comparison between the non-ADHD and ADHD groups for each characteristic.†LC9 risk factor count was calculated as the sum of 9 binary LC9 risk factors, with higher count indicating greater overall risk burden.Abbreviation: ADHD, Attention deficit and hyperactivity disorder. Overall LC9 risk burden and distribution of LC9 risk factors at Wave V according to ADHD status. aModel 1: Unadjusted Model. bModel 2: Adjusted for age, sex, race, highest education level, annual family income and alcohol consumption. *The P-values represent the statistical comparison between the non-ADHD and ADHD groups for each characteristic. †LC9 risk factor count was calculated as the sum of 9 binary LC9 risk factors, with higher count indicating greater overall risk burden. Abbreviation: ADHD, Attention deficit and hyperactivity disorder. Table 3Cumulative cardiovascular risk burden categories and multinomial logistic regression analyses according to ADHD status at Wave V.Table 3 dummy alt textCumulative LC9 risk factors, N (%)Non-ADHD (N = 3331)ADHD (N = 168)OR (95 % CI)Model 1Model 20–1 risk factors1168 (31.39)26 (12.08)1.00 (reference)1.00 (reference)2 risk factors858 (25.95)38 (23.32)2.34 (1.11–4.93)2.80 (1.23–6.38)3 risk factors617 (19.53)42 (23.70)3.15 (1.55–6.40)4.09 (1.96–8.56)≥4 risk factors688 (23.13)62 (40.90)4.60 (2.31–9.15)4.51 (2.06–9.84)aModel 1: Unadjusted Model.bModel 2: Adjusted for age, sex, race, highest education level, annual family income and alcohol consumption.*The cumulative LC9 risk factor analyses were restricted to participants with complete data on all 9 LC9 components (Non-ADHD: n = 3331; ADHD: n = 168). Percentages are weighted percentages.†Odds ratios were estimated using multinomial logistic regression, with 0–1 LC9 risk factors as the reference category.Abbreviation: ADHD, Attention deficit and hyperactivity disorder; OR, odds ratio. Cumulative cardiovascular risk burden categories and multinomial logistic regression analyses according to ADHD status at Wave V. aModel 1: Unadjusted Model. bModel 2: Adjusted for age, sex, race, highest education level, annual family income and alcohol consumption. *The cumulative LC9 risk factor analyses were restricted to participants with complete data on all 9 LC9 components (Non-ADHD: n = 3331; ADHD: n = 168). Percentages are weighted percentages. †Odds ratios were estimated using multinomial logistic regression, with 0–1 LC9 risk factors as the reference category. Abbreviation: ADHD, Attention deficit and hyperactivity disorder; OR, odds ratio. Associations between ADHD and each LC9 components We examined the associations between ADHD and individual components of Life’s Crucial 9 (LC9) at Wave V. After adjustment for demographic covariates, individuals with ADHD had significantly higher odds of four LC9 components—current smoking, sleep problems, hyperlipidemia, and anxiety and/or depression—compared with those without ADHD. After further adjustment for anxiety and depression, ADHD remained significantly associated with current smoking (OR, 1.66 [95 % CI, 1.26–2.18]) and sleep problems (OR, 1.41 [95 % CI, 1.09–1.82]) (Fig. 1 and Table S2). Notably, supplementary cross-sectional analyses at Wave IV demonstrated associations in the same direction, supporting the robustness of the prospective findings (Tables S3 and S4).Fig. 1Associations between ADHD and individual LC9 risk factors. Odds ratios (ORs) and 95 % confidence intervals (CIs) for the associations between attention-deficit/hyperactivity disorder (ADHD) and each component of Life’s Crucial 9 (LC9) at Wave V. Logistic regression models compare participants with versus without ADHD. Model 1–3 represent sequentially adjusted models as described in the Methods. Psychological health problem refers to anxiety and/or depression. The vertical dashed line indicates the null value (OR = 1.0). Results correspond to those presented in Table S2.Fig 1 dummy alt text Associations between ADHD and individual LC9 risk factors. Odds ratios (ORs) and 95 % confidence intervals (CIs) for the associations between attention-deficit/hyperactivity disorder (ADHD) and each component of Life’s Crucial 9 (LC9) at Wave V. Logistic regression models compare participants with versus without ADHD. Model 1–3 represent sequentially adjusted models as described in the Methods. Psychological health problem refers to anxiety and/or depression. The vertical dashed line indicates the null value (OR = 1.0). Results correspond to those presented in Table S2. Joint-category analyses by psychological health status Joint-category analyses were conducted to evaluate LC9 risk factors according to combined ADHD and psychological health status. Compared with participants without ADHD and without anxiety and/or depression, those without ADHD but with anxiety and/or depression had higher odds of several adverse outcomes, including current smoking, low physical activity, obesity, sleep problems, hyperlipidemia, hypertension, and diabetes. Participants with both ADHD and anxiety and/or depression showed the most adverse profiles for current smoking (OR, 2.38 [95 % CI, 1.62–3.48]), sleep problems (OR, 1.94 [95 % CI, 1.34–2.81]), hyperlipidemia (OR, 2.27 [95 % CI, 1.27–3.96]), and hypertension (OR, 1.74 [95 % CI, 1.01–2.98]) (Table 4). Similar patterns were observed in supplementary analyses of ADHD in combination with anxiety and depression separately (Tables S5 and S6). Formal ADHD × psychological distress interaction identified evidence of effect modification only for current smoking (interaction coefficient: −0.73; P for interaction = 0.02), whereas no statistically significant interaction was observed for the other outcomes (Table S7).Table 4Association between ADHD and psychological health diagnosis and “Life’s Crucial 9” risk factors of cardiovascular disease in Wave V.Table 4 dummy alt textNon-ADHD & No-psychological problem (N=7889)Non-ADHD & Psychological problem (N=2517)P valueADHD & No-psychological problem (N=244)P valueADHD & Psychological problem (N=246)P valueCurrent SmokingNo. of cases/total participants3209/76401384/2449155/24080/160Model 11.00 (reference)1.97(1.74–2.28)<.0012.10 (1.45–3.04)<.0012.28 (1.54–3.38)<.001Model 21.00 (reference)1.83 (1.58–2.11)<.0012.16 (1.47–3.19)<.0012.38 (1.62–3.48)<.001Low physical activityNo. of cases/total participants436/6859164/221740/19621/218Model 11.00 (reference)1.89 (1.51–2.35)<.0010.93 (0.38–2.24)0.871.99(1.17–3.38)0.01Model 21.00 (reference)1.63 (1.30 = 2.06)<.0010.82 (0.40–1.44)0.571.77 (1.00–3.14)0.05Dietary habitsNo. of cases/total participants592/6104285/191916/18733/187Model 11.00 (reference)1.14 (0.88–1.47)0.341.23 (0.70–2.17)0.461.16(0.61–2.14)0.64Model 21.00 (reference)1.03 (0.78–1.36)0.821.20 (0.65–2.17)0.561.07 (0.55–2.06)0.84ObesityNo. of cases/total participants1540/3499658/131261/11170/129Model 11.00 (reference)1.40 (1.16–1.68)0.0041.66 (0.97–2.84)0.061.70 (1.01–2.84)0.04Model 21.00 (reference)1.25 (1.05–1.52)0.021.41 (0.80–1.52)0.191.47 (0.84–2.57)0.25Sleep problemNo. of cases/total participants3451/78811320/2514122/243125/244Model 11.00 (reference)1.50 (1.32–1.70)<.0011.60 (1.12–2.29)0.011.91 (1.33–2.74)<.001Model 21.00 (reference)1.44 (1.26–1.65)<.0011.61 (1.11–2.33)0.101.94 (1.34–2.81)0.005HyperlipidemiaNo. of cases/total participants475/3564263/133823/11631/132Model 11.00 (reference)1.81 (1.44–2.26)<.0011.81(1.01–3.17)0.042.44 (1.42–4.27)0.001Model 21.00 (reference)1.67 (1.31–2.12)<.0011.67 (0.83–3.33)0.152.27 (1.27–3.96)0.007HypertensionNo. of cases/total participants992/3566455/133849/11648/132Model 11.00 (reference)1.55(1.27–1.90)<.0011.98 (1.15–3.43)0.021.85 (1.10–3.11)0.02Model 21.00 (reference)1.41 (1.15–1.74)0.0011.49 (0.86–2.62)0.151.74 (1.01–2.98)0.04DiabetesNo. of cases/total participants269/3564158/133810/11615/132Model 11.00 (reference)1.72(1.29–2.30)<.0011.57 (0.48–5.07)0.481.08 (0.48–2.41)0.84Model 21.00 (reference)1.75 (1.28–2.38)0.0040.68 (0.29–1.60)0.370.87 (0.35–2.16)0.75aModel 1: Adjusted for age, sex and race.bModel 2: Adjusted for age, sex, race, highest education level, annual family income and alcohol consumption.Abbreviation: ADHD, Attention deficit and hyperactivity disorder; OR, odds ratio. Table 4Association between ADHD and psychological health diagnosis and “Life’s Crucial 9” risk factors of cardiovascular disease in Wave V.Table 4 dummy alt textNon-ADHD & No-psychological problem (N=7889)Non-ADHD & Psychological problem (N=2517)P valueADHD & No-psychological problem (N=244)P valueADHD & Psychological problem (N=246)P valueCurrent SmokingNo. of cases/total participants3209/76401384/2449155/24080/160Model 11.00 (reference)1.97(1.74–2.28)<.0012.10 (1.45–3.04)<.0012.28 (1.54–3.38)<.001Model 21.00 (reference)1.83 (1.58–2.11)<.0012.16 (1.47–3.19)<.0012.38 (1.62–3.48)<.001Low physical activityNo. of cases/total participants436/6859164/221740/19621/218Model 11.00 (reference)1.89 (1.51–2.35)<.0010.93 (0.38–2.24)0.871.99(1.17–3.38)0.01Model 21.00 (reference)1.63 (1.30 = 2.06)<.0010.82 (0.40–1.44)0.571.77 (1.00–3.14)0.05Dietary habitsNo. of cases/total participants592/6104285/191916/18733/187Model 11.00 (reference)1.14 (0.88–1.47)0.341.23 (0.70–2.17)0.461.16(0.61–2.14)0.64Model 21.00 (reference)1.03 (0.78–1.36)0.821.20 (0.65–2.17)0.561.07 (0.55–2.06)0.84ObesityNo. of cases/total participants1540/3499658/131261/11170/129Model 11.00 (reference)1.40 (1.16–1.68)0.0041.66 (0.97–2.84)0.061.70 (1.01–2.84)0.04Model 21.00 (reference)1.25 (1.05–1.52)0.021.41 (0.80–1.52)0.191.47 (0.84–2.57)0.25Sleep problemNo. of cases/total participants3451/78811320/2514122/243125/244Model 11.00 (reference)1.50 (1.32–1.70)<.0011.60 (1.12–2.29)0.011.91 (1.33–2.74)<.001Model 21.00 (reference)1.44 (1.26–1.65)<.0011.61 (1.11–2.33)0.101.94 (1.34–2.81)0.005HyperlipidemiaNo. of cases/total participants475/3564263/133823/11631/132Model 11.00 (reference)1.81 (1.44–2.26)<.0011.81(1.01–3.17)0.042.44 (1.42–4.27)0.001Model 21.00 (reference)1.67 (1.31–2.12)<.0011.67 (0.83–3.33)0.152.27 (1.27–3.96)0.007HypertensionNo. of cases/total participants992/3566455/133849/11648/132Model 11.00 (reference)1.55(1.27–1.90)<.0011.98 (1.15–3.43)0.021.85 (1.10–3.11)0.02Model 21.00 (reference)1.41 (1.15–1.74)0.0011.49 (0.86–2.62)0.151.74 (1.01–2.98)0.04DiabetesNo. of cases/total participants269/3564158/133810/11615/132Model 11.00 (reference)1.72(1.29–2.30)<.0011.57 (0.48–5.07)0.481.08 (0.48–2.41)0.84Model 21.00 (reference)1.75 (1.28–2.38)0.0040.68 (0.29–1.60)0.370.87 (0.35–2.16)0.75aModel 1: Adjusted for age, sex and race.bModel 2: Adjusted for age, sex, race, highest education level, annual family income and alcohol consumption.Abbreviation: ADHD, Attention deficit and hyperactivity disorder; OR, odds ratio. Association between ADHD and psychological health diagnosis and “Life’s Crucial 9” risk factors of cardiovascular disease in Wave V. aModel 1: Adjusted for age, sex and race. bModel 2: Adjusted for age, sex, race, highest education level, annual family income and alcohol consumption. Abbreviation: ADHD, Attention deficit and hyperactivity disorder; OR, odds ratio. Sex-stratified analyses of LC9 risk factors Sex-stratified analyses were conducted to assess whether associations between ADHD and LC9 risk factors differed between men and women. In both sexes, ADHD was associated with higher odds of current smoking and psychological distress. The association with current smoking appeared numerically stronger among men (OR, 1.69 [95 % CI, 1.17–2.45]) and the association with psychological distress appeared stronger among women (OR, 4.02 [95 % CI, 2.65–6.10]). ADHD was also associated with sleep problems among women (OR, 1.85 [95 % CI, 1.20–2.85]), while the association with hyperlipidemia appeared stronger among men (OR, 1.74 [95 % CI, 1.01–3.13]) (Tables S8 and S9). Formal testing for ADHD × sex interaction yielded no statistically significant results in either Model 2 or Model 3 (all P for interaction > 0.05). Sensitivity analysis In sensitivity analyses, we additionally adjusted for available baseline (Wave IV) measures of specific LC9 components to account for pre-existing risk levels. The overall pattern of results remained broadly similar, although some associations were attenuated (Table S10). Specifically, sleep problems remained significantly associated with ADHD after baseline adjustment, whereas the association with current smoking was weakened. Modified Poisson regression with robust variance yielded directionally consistent results, although the estimated relative risks were smaller in magnitude than the corresponding odds ratios for common outcomes (Table S11). Overall, these analyses did not materially alter the interpretation of the primary findings, further supporting the robustness of the observed associations. Discussion In this nationally representative cohort of young adults in the U.S., ADHD was prospectively associated with a higher cumulative burden of LC9 risk factors, specifically with increased odds of cigarette smoking, sleep problems, and psychological distress at Wave V. Joint-category analyses showed that participants with comorbid ADHD and psychological distress exhibited the most adverse profiles for several outcomes, most notably smoking, sleep problems, hyperlipidemia, and hypertension. Notably, young adults with ADHD exhibited a distinct shift toward higher categories of cumulative LC9 risk burden relative to their non-ADHD peers. Adjusted multinomial regression models demonstrated that ADHD was associated with significantly higher odds of belonging to the 2, 3, and ≥4 risk factor categories, compared with the 0–1 category. These findings suggest that the impact of ADHD on cardiovascular health extend beyond individual LC9 components, involving a broader accumulation of adverse cardiovascular and psychological burdens in early adulthood. To our knowledge, this is the first study to examine the association between ADHD and cumulative LC9 risk factor burden in young adults. Our findings are broadly consistent with prior studies linking ADHD to smoking and sleep disturbance [22,23]. By contrast, we did not observe clear overall associations with physical inactivity or unhealthy dietary habits, despite previous reports suggesting that individuals with ADHD may be less likely to meet recommended physical activity levels or more likely to consume unhealthy foods [24,25]. Furthermore, existing evidence for cardiometabolic outcomes, such as obesity and hypertension in ADHD, remains inconsistent [19,[26], [27], [28]]. In the present study, the associations of ADHD with obesity and hyperlipidemia were attenuated after full adjustment, and no clear association with diabetes was observed. Taken together, these findings suggest that in young adulthood, ADHD is more consistently associated with smoking, sleep, and psychological distress than with all cardiometabolic outcomes uniformly. This indicates that associations with cardiometabolic outcomes may be more modest and highly sensitive to covariate adjustment and analytic specification. In sex-stratified analyses, although some associations appeared to differ numerically between men and women, formal ADHD × sex interaction tests yielded no statistically significant results. Consequently, these subgroup-specific patterns should be interpreted as descriptive rather than as evidence of sex-specific effect modification. Notably, ADHD is frequently comorbid with psychological health issues such as depression, anxiety, and mood disorders [29,30]. Consistent with this, our joint-category analyses showed that co-occurring ADHD and psychological problems identified a subgroup with the highest clinical burden. While formal interaction evidence was limited, these patterns underscore the clinical necessity of integrating mental health assessments into cardiovascular risk stratification for this population. This seemingly modest improvement in predictive performance when additionally adjusting for psychological health warrants a nuanced interpretation. One plausible interpretation is that psychological factors and ADHD may function as upstream determinants of later established cardiometabolic risk factors [31]. Under this framework, the attenuated estimates in fully adjusted models do not imply that ADHD is a negligible risk factor; rather, this pattern suggests that its impact may be partially mediated through downstream pathways, including psychological distress, sleep disturbances, and adverse health behaviors. These interpretations remain tentative and require confirmation in future longitudinal studies with repeated measures and causal mediation analyses. Nevertheless, our longitudinal study design underscores that accounting for the temporal order of occurrence reveals a critical role for ADHD in the early development of an unfavorable cardiovascular profile, thereby justifying the integration of mental health into comprehensive cardiovascular prevention strategies. Several mechanisms may underlie the observed associations, although our data do not permit direct mechanistic inference. Core ADHD symptoms, particularly impulsivity, may contribute to unhealthy behaviors such as smoking and poor sleep hygiene [31,32]. Shared neurobiological pathways, including dysregulated dopaminergic signaling and altered circadian regulation, may also contribute to both ADHD pathophysiology and behavioral dysregulation [33,34]. In addition, co-occurring anxiety and depression may amplify cardiovascular risk through behavioral and physiological stress pathways [35]. ADHD-related executive dysfunction may further impair self-regulation, reinforcing unhealthy behaviors and psychological distress [36]. We acknowledge that our application of the LC9 framework focuses on pathological thresholds rather than the continuous spectrum of cardiovascular health (CVH) traditionally emphasized by the American Heart Association (AHA). The AHA’s population-based approach aims to capture sub-clinical risk elevations to promote ideal health. However, in the context of young adults with ADHD—a group often characterized by significant diagnostic and treatment gaps—the use of categorical risk factors provides a pragmatic assessment of clinical burden. While this binary approach may overlook individuals with moderate risk elevations and potentially lead to non-differential misclassification, it offers a clear and interpretable metric for the accumulation of established cardiovascular risk factors in this high-risk neurodevelopmental population. This study has several strengths, including the use of a large, nationally representative cohort with diverse racial and ethnic representation, the application of an LC9-based framework across multiple behavioral, psychological, and cardiometabolic domains, and the inclusion of sensitivity and interaction analyses that strengthened the robustness and interpretability of the findings. However, several limitations should be acknowledged. First, ADHD status was determined from a self-reported history of clinician diagnosis at, which may be subject to misclassification bias. However, such misclassification is likely to be non-differential with respect to cardiometabolic outcomes in this prospective design. Theoretically, non-differential misclassification tends to bias toward the null. Therefore, the association observed in this study between ADHD and cardiovascular risk factors might be underestimations of the true effect sizes. Second, due to data constraints in the Add Health study, we were unable to assess specific clinical characteristics of ADHD, such as the age of onset, symptom severity, and the total duration of the disorder. The absence of such granular data limits our capacity to determine how the specific clinical trajectory of ADHD influences cardiovascular risk profiles. Third, Common ADHD treatments, particularly stimulants, are known to be associated with increased heart rate, elevated blood pressure, and potential sleep problems [35,37]. However, the Add Health study lacks detailed information on ADHD medication history, which limits our ability to account for the potential confounding effects of medication on cardiometabolic outcomes. Future studies incorporating pharmacy records are needed. Fourth, while this study utilized a prospective design with ADHD assessed at Wave IV and LC9 factors at Wave V, the exact temporal ordering of incidence remains an inherent limitation. Some cardiovascular risk factors, such as unhealthy dietary habits or obesity, may develop early in life before the reported ADHD diagnosis. Without definitive data on the timing of onset for each LC9 component, we cannot exclude the possibility of reverse causality or bidirectional associations. Finally, the relatively young age of this cohort also means these associations reflect early-life risk accumulation rather than clinical cardiovascular events, which may limit generalizability to older populations. In conclusion, ADHD in young adulthood was primarily associated with higher burdens of smoking, sleep, and psychological problems, and a greater overall accumulation of LC9 risk factors. These findings, particularly the heightened risk in those with psychological comorbidities, suggest that ADHD is an important marker of early-life cardiovascular risk. This supports the need for integrated risk assessment; however, further mechanistic and interventional studies are required to determine if addressing ADHD symptoms directly improves cardiovascular outcomes. Declaration of generative AI and AI-assisted technologies in the writing process During the preparation of this manuscript, the authors used ChatGPT (OpenAI) only for language polishing and improving readability. All AI-assisted edits were carefully reviewed and revised by the authors, who take full responsibility for the final content. Funding This study was funded by the 10.13039/501100001809National Natural Science Foundation of China (32241010) and the Young Researchers’ Innovative Funding, 10.13039/501100009076University of Science and Technology of China (ID: K9100000060). CRediT authorship contribution statement Na Li: Conceptualization, Methodology, Investigation, Data curation, Software, Formal analysis, Writing – original draft. Meiyuan Chen: Conceptualization, Methodology. Sidong Li: Conceptualization, Methodology, Software, Formal analysis, Data curation. Yang Du: Conceptualization, Validation. Yuxiao Wu: Conceptualization, Validation. Xiaodong Zang: Conceptualization, Methodology, Validation. Ruishu Zhang: Visualization. Zihan Zhang: Visualization. Buyun Liu: Conceptualization, Validation. Wei Bao: Conceptualization, Supervision. Guifeng Xu: Conceptualization, Supervision, Resources, Writing – review & editing, Project administration, Funding acquisition. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.