Challenges and opportunities for the prevention and treatment of cardiovascular disease among young adults

Discussion and Emerging Themes

Workshop presentations were organized into multiple sessions: 

• Review of the current knowledge about young adult CVH and the development of risk factors, primarily from contemporary epidemiologic studies and cohorts including:  the National Health and Nutrition Examination Survey (NHANES), National Longitudinal Study of Adolescent to Adult Health (Add Health), Coronary Artery Risk Development in Young Adults (CARDIA), Pathobiological Determinants of Atherosclerosis in Youth (PDAY) study, the Treatment Options for Type 2 Diabetes in Adolescents and Youth (TODAY) study, and the International Childhood Cardiovascular Cohort (i3C) Consortium;
• Discussion of the lifestyle, psychological, and social contributors to CVH and CVD risk of young adults;
• Discussion of the occurrence of CVD outcomes in young adulthood and the association of co-occurring conditions such as diabetes and pregnancy with CVD outcomes;
• Review of behavioral and pharmacologic interventions targeting prevention or delay of CVD outcomes, including issues related to study design and outcome ascertainment.   

Several themes emerged from the discussion and are presented in detail below: 

Trends and Transitions

The transition from adolescence to young adulthood is a time of dramatic psychological, social, and biological development that poses unique challenges and opportunities for the prevention of CVD.  For example, this life stage is characterized by increased autonomy over decisions that impact health, and the influence of peers and romantic partners becomes more prominent. Young adults may have unpredictable schedules and shifting priorities due to job and/or school transitions. This results in difficulties meeting guidelines for healthy sleep, physical activity, and diet.  Young adults have a stronger interest in shared decision making with health care providers, but job and/or school transitions may result in seeing multiple health care providers and/or precipitate gaps in health insurance coverage, which may contribute to a delay in engagement in care overall and particularly for preventive care.  Silos across medical specialties and lack of coordination across providers, such as from pediatric to adult care or between obstetricians and primary care providers for young adult childbearing women complicate care.

Societal changes that are specific to the “modern” cohort of contemporary 21st century young adults must also be considered.[1] Modern day young adults are more inclined to desire autonomous, easy access to medical care facilitated by the use of online and mobile technology (e.g., 24 hour “minute clinics” and pharmacies, online scheduling, habit tracking and behavior change apps). Modern cohorts of young adults have higher rates and earlier onset of several key CVD risk factors (such as obesity, hypertension, and type 2 diabetes) compared to prior cohorts, despite the lower rates of some other CVD risk factors (specifically cigarette smoking and dyslipidemia).[2],[3],[4] Changing policy trends have resulted in more young adults with insurance coverage, thus reducing this barrier in access to medical care.[5]

Transitions and recent trends impacting today’s young adults should inform the development and implementation of interventions for CVD prevention and management.  The ubiquity of connected technology and accessible mobile applications, particularly among young adults, represent opportunities but also present challenges for understanding how these might be used effectively for advancing CVD health.  Although these approaches may increase uptake and engagement-- especially for tracking behaviors such as diet, sleep, and exercise--the rapid pace of technology development may outstrip the pace of research. In addition, the absent or delayed engagement of young adults in medical care may challenge the potential usefulness of electronic medical data for assessing population health and the conduct of pragmatic trials.

Resilience

The young adulthood period may be particularly well suited for studying both biological and psychological resilience, defined as well-being in the face of risk or adversity. Contributors to resilience may be important for understanding many patterns of variation in health outcomes but have not been well characterized.  In this life stage, individuals face the challenges of forming identity and achieving financial independence. This period also provides an opportunity to build “reserve capacity” and render one less susceptible to exposure to and/or accumulation of psychological, social, and environmental risks into later adulthood.  There is some research in this area, with interventions based in the Positive Youth Development framework[6] and mindfulness programs[7] showing improved outcomes such as reductions in teen pregnancy[8] and depression[9]; however, there has been little study of their impact on CV outcomes.[10] 

Furthermore, it would be important to understand how differences in resilience may contribute to disparities in CV health.[11]  For example, there are racial differences in the protective effect of higher socioeconomic status on CV health, with SES conferring less of a protective effect on the CV health of higher SES young black adults compared to young white adults.  This may be due to the “costs” of social mobility (e.g., cultural and/or social isolation, striving, or being too busy for prevention and medical care).

Risk ascertainment and risk perception

The perception of risk for CVD among young adults may be consistent with their stage in psychological and social development. There is a notion that cognitive biases lead us to make apparently irrational decisions in terms of life-long health consequences: for example, the rule-of-thumb that one cannot plan for events occurring beyond twice one’s age.[12]  Thus, the perceived urgency for adopting preventive and risk reduction behaviors to mitigate the risk of CVD outcomes may be low for early young adulthood (e.g., late adolescence through 20s); whereas, for early midlife (e.g., 30s) this cognitive style may support an inclination toward adopting health promoting behaviors as a “mid-course correction”.  In addition, the younger age group may be more prone to a perception that having a family history of CVD means their personal risk is genetic and not modifiable.  Recent evidence from NHANES shows there is poor awareness of the presence of CV risk factors such as hypertension and diabetes among adults age 20-39.[13],[14]  On the other hand, increasing perception of risk may not be the correct approach for motivating young adults to be oriented towards CVD prevention.  Developmentally, young adults incline towards positive, growth oriented, achievement oriented, and prosocial motivations which may be more activating for prevention.  Research is needed to understand how to motivate young adults’ adherence to needed behavior changes, medication, or doctor’s visits.  Building resilience may guard against negative factors influencing young adult health and behavior. A greater understanding of how to harness those developmentally appropriate attributes of young adults may promote healthful behavior change.

Primary care providers are also subject to barriers in assessing risk among young adult patients. Providers may be hesitant to “label” younger individuals, but workshop participants suggest that having a specific diagnosis often contributes to increased awareness and interest in prevention and/or treatment. It is well known that hypertension awareness is low among US adults generally and this is also true in the young adult period (74.7% age 18-39 years versus 81.9% age 40-59 years versus 88.4% ≥60 years). Even among those aware, treatment (50.0% versus 70.3% versus 83.0%) and control (40.2% versus 56.7% versus 54.4%) rates are also low.[15] Perception among providers that CVD risk is far in the future and therefore control of these risk factors is less urgent may be a factor contributing to these rates and studies suggest less rapid follow-up of elevations in blood pressure.  In one study, only 56% of young adults age 18-24 years with multiple elevated blood pressure readings in the electronic health record of a health system received a hypertension diagnosis within 48 months of follow-up.[16] Furthermore, masked hypertension, defined as a normal clinic blood pressure with elevated ambulatory levels, may represent another scenario of under-diagnosis for young adults. There are few studies of ambulatory blood pressure monitoring among young adults, but masked hypertension prevalence may be somewhere between 8 and 20% based on current available data of youth and adults.[17]  

Although young adults are at lower overall risk for CVD events in the short term, as a result of accumulating CVD risk this is not uniformly true and a higher risk subset of young adults exist (e.g., Blacks and American Indians, childhood cancer survivors, type 1 and type 2 diabetics, individuals with familial hypercholesterolemia and chronic kidney disease) who may be targets for more aggressive and timely interventions to modify risk.[18]  More research is needed in this area to inform providers about prioritizing modification of CVD risk in this population.  Another potential barrier is the lack of sufficient evidence to inform the optimal timing and effectiveness of initial and follow-up assessment for CVD risk factors and subclinical CVD among young adults.  For example, although the Pathobiological Determinants of Atherosclerosis in Youth (PDAY) risk score has been validated to predict atherosclerotic lesions among young adults and middle-aged subjects[19], to our knowledge the PDAY score is not routinely used in clinical practice.  The absence of evidence to inform clinical guidelines targeting young adults results in under-recognition of risk and undertreatment by providers. The 2013 AHA/ACC adult risk assessment guidelines provided calculators for lifetime CVD risk assessment starting at middle age (minimum age 40 years) and emphasized use of 10-year CVD risk to guide statin treatment decisions.  The lack of risk assessment guidelines for young adults is partly a product of sparse epidemiological and clinical trials evidence for this group.[20]  Additional gaps include evidence for sequential versus simultaneous interventions in multiple CVD risk factors, as well as improved non-invasive assessments of lifetime risk that may help guide approaches to prioritizing interventions.

Risk Trajectories

Recent research has focused on CVD risk factor trajectories and development of CVD outcomes.  Obesity, hypertension, and smoking are the risk factors most consistently and markedly associated with CVD outcomes among young adults.[21] Epidemiologic data on population trends in risk factor prevalence and incidence show that each risk factor has a unique trajectory for development during the life course.  For modern cohorts of young adults, risk factor accumulation is occurring at earlier ages and developmental stages, which may have implications on the timing and types of interventions that should be considered.  For example, obesity onset is occurring at younger ages (8 or 9 years old), whereas smoking may occur in late adolescence and early adulthood.  The earlier onset of obesity is occurring concomitantly with a population trend toward earlier puberty; the relationship of pubertal timing with childhood and adolescent obesity and their association with adult CVD risk has not been fully elucidated.[22]

Risk behaviors are interrelated, further complicating understanding of trajectories and approaches to intervention.  For example, effectively treating smoking causes overeating and weight gain for 80%, and treating the two simultaneously can be overwhelming and cause dropout. Furthermore, simultaneously treating some risk factors while still needing to prevent others as well as not knowing which (or how many) risk behaviors to address adds complexity to comprehensive management of CVD risk. This complexity may impact the sense of urgency to address concurrent issues from both the patient and provider perspective. Interventions for a given risk factor may have a different impact on one’s baseline risk, i.e., some interventions may result in risk regression rather than just stabilization or slowed progression.  For example, weight loss and smoking cessation can have one return to baseline risk.[23] However, hypertension treatment to normal blood pressure later in life still has residual risk for CVD outcomes;[24] by the time treatment is initiated some subclinical organ damage may have already occurred. On the other hand, the implications of initiating medications for treatment of hypertension and hyperlipidemia in early adulthood is not clear, as experience with decades-long medication is limited.  A third possibility is that the usual risk factor management paradigm will not suffice for young adults.  Specifically, it is unknown whether lifetime medication therapy is necessary, or if early treatment, over a discreet period could result in prolonged residual risk factor control and delay or perhaps even eliminate the need to reinitiate medication.

Lastly, more research is needed on the role of social, environmental, behavioral, and psychological attributes on predicting risk trajectories, and identifying which of these attributes may be modified and may change one’s risk trajectory.  Such evidence would further justify the expansion of risk assessment to include psychosocial risk (e.g., poverty, adverse childhood experiences, incarceration).

Women’s health and gender differences

It is well established that men and women have different trajectories for acquisition of CV risk factors and CV outcomes over the life course; the young adult period presents unique challenges and opportunities for understanding these differences.  Obese and overweight young adult women may be more susceptible to body image concerns, which may result in delays in medical care engagement due to a desire to lose weight before seeing a health care provider.[25]  However, young adult women are more likely than men to have a preventive visit with a health care provider due, in part, to reproductive health needs.[26]  

The relationship of pregnancy and childbirth with cardiovascular disease risk is an area for continued investigation.[27]  It is not clear if pregnancy simply unmasks underlying risk due to the changes in blood volume and vascular compliance, or if it confers additional risk.  There is some evidence to suggest that there is a residual elevated CV risk even after controlling for usual pre-pregnancy risk factors.  Additional research is needed to understand the timing (i.e., preconception, during pregnancy, postpartum) and type of interventions that may reduce pregnancy-related CVD risk.

Outcome Ascertainment

Participants recognized the challenges in quantifying the effect of interventions targeting young adults within short follow-up periods typical of clinical trials (4-5 years), particularly if relying upon CVD event endpoints (e.g., myocardial infarction) and/or all-cause mortality.  From a prevention standpoint, subclinical CVD endpoints may be more reasonable for evaluating interventions.  More research is needed to determine the best measures of subclinical CVD that address issues related to measurement consistency and costs.  Pooled analyses or the use of synthetic cohorts may be needed.  Existing cohorts of young adults have varying and limited measures of subclinical CVD. For example, the individual cohorts within the i3C Consortium had too few participants with subclinical measures for cohort-specific age stratified risk analyses; however, when combined in an analysis of pooled data, the ages when risk factor associations with subclinical disease measures are evident were identified. Existing cohorts may also lack sufficient assessment of key risk factors such as patterns of diet, sleep, and physical activity. 

Computer simulation methods that forecast downstream CVD outcomes based on clinical trial and observational evidence may identify important areas for research and support intervention trial design. In addition, it is essential to systematically evaluate the design and impact of recruitment and retention strategies to ensure that the collection and interpretation of outcome data from multi-year longitudinal cohort studies and intervention trials are of highest quality in the future. 

Summary and overall conclusions

Participants expressed enthusiasm for the critical need to focus on young adult CVH and the urgency of this focus given the demographic trends in the US and the trend in increasing CVD risk.  While the scope of the discussion was broad and focused on multiple types of CVD risk factors, areas of most urgent need included:

1. Methodologic – given the challenges of studying young adults (enrollment and retention, long time to accrual of outcomes), attention should be focused on improved methodology for study and interpretation of results from existing data sources (e.g., linked cohorts, electronic health records, etc.).
2. Observational studies – more information on the timing of risk and the timing of intervention across individual risk factors and multiple risk factors is important for continuing to guide intervention studies and communication of risk to young adults and providers.  Additionally, identifying those higher risk individuals who may be appropriate for more aggressive intervention should also be a focus.
3. Interventions – more is needed to understand how to harness new technologies to enhance interventions and to harness the developmentally appropriate aspects of intervention in this period (positive motivated affect, resilience) to improve prevention.

Publication Plans

The organizers for the workshop plan to publish a report from the meeting with a summary of the discussion and recommendations in a peer reviewed journal.

NHLBI Contact

Nicole Redmond, MD, PhD, MPH, FACP
Medical Officer, Clinical Applications and Prevention Branch, Division of Cardiovascular Sciences, NHLBI
nicole.redmond@nih.gov

Meeting Participants

Chair

• Kirsten Bibbins-Domingo, PhD, MD, MAS
• Lee Goldman, MD Endowed Chair in Medicine and Professor of Epidemiology and Biostatistics and Medicine, University of California-San Francisco (UCSF)

Members

• Norrina Allen, PhD, MPH, FAHA, Associate Professor, Department of Preventive Medicine, Feinberg School of Medicine
• Fida Bacha, MD, FAAP, Associate Professor of Pediatrics, Texas Children’s Hospital/Baylor College of Medicine
• Trudy L. Burns, MPH, PhD, Professor, College of Public Health, Carver College of Medicine, and College of Nursing, University of Iowa
• Janet M. Catov, PhD, MS, Associate Professor, Department of Obstetrics, Gynecology & Reproductive Sciences and the Department of Epidemiology, University of Pittsburgh
• Samuel S. Gidding, MD, Chief, Pediatric Cardiology, Division of Pediatric Cardiology, Department of Pediatrics Cardiology, Nemours Cardiac Center/ Alfred I. duPont Hospital for Children
• Holly Gooding, MD, MSc, Assistant Professor, Harvard Medical School/Boston Children’s Hospital
• Michael A. Grandner, PhD, Assistant Professor, Departments of Psychiatry and Medicine, College of Medicine and Department of Psychology College of Science, University of Arizona
• Heather M. Johnson, MD, MS, FAHA, Associate Professor of Medicine, Division of Cardiovascular Medicine, University of Wisconsin School of Medicine & Public Health
• Michaela Kiernan, PhD, Senior Research Scientist, Stanford Prevention Research Center, Department of Medicine, Stanford University School of Medicine
• Tené T. Lewis, PhD, Associate Professor, Department of Epidemiology, Rollins School of Public Health, Emory University
• Karen Matthews, PhD, Distinguished Professor of Psychiatry and Professor of Epidemiology, Psychology, and Clinical and Translational Science at the University of Pittsburgh
• Maureen Monaghan, PhD, Assistant Professor of Psychology and Behavioral Health, Children’s National Health System and George Washington University School of Medicine
• Andrew E. Moran, MD, MPH, Assistant Professor of Medicine, Division of General Medicine, Department of Medicine, Columbia University 
• Kathleen Mullan Harris, PhD, James E. Haar Distinguished Professor of Sociology and Director of the National Longitudinal Study of Adolescent to Adult Health (Add Health), University of North Carolina at Chapel Hill
• Jennifer G. Robinson, MD, MPH, Professor, Departments of Epidemiology and Medicine (Division of Cardiology), and Director of the Prevention Intervention Center at the University of Iowa
• Bonnie Spring, PhD, ABPP, Professor of Preventive Medicine, Psychology, Psychiatry & Public Health; Director, Center for Behavior and Health, Institute for Public Health & Medicine (IPHAM), Northwestern University Feinberg School of Medicine
• Deborah Tate, PhD, Professor, Department of Health Behavior and Department of Nutrition, Gillings School of Global Public Health, University of North Carolina at Chapel Hill

NHLBI Staff Speakers and Moderators

• David C. Goff, Jr., MD, PhD, FACP, FAHA , Director, Division of Cardiovascular Sciences, National Heart, Lung, and Blood Institute
• Rebecca Campo, PhD, MS, Program Director, Clinical Applications and Prevention Branch, Division of Cardiovascular Sciences, NHLBI
• Charlotte Pratt, PhD, RD, FAHA, Acting Deputy Branch Chief, Clinical Applications and Prevention Branch, Division of Cardiovascular Sciences, NHLBI
• Jared Reis, PhD, Program Director, Epidemiology Branch, Division of Cardiovascular Sciences, NHLBI
• Catherine (Kate) Stoney, PhD, Acting Deputy Branch Chief, Clinical Applications and Prevention Branch, Division of Cardiovascular Sciences, NHLBI