Incorporating Cardiovascular Risk Assessment into Adolescent Reproductive Health and Primary Care Visits

Abstract

Study Objective

This study aimed to evaluate the usability and feasibility of incorporating a cardiovascular risk assessment tool into adolescent reproductive health and primary care visits.

Methods

We recruited 60 young women ages 13–21 years to complete the HerHeart web-tool in two adolescent clinics in Atlanta, GA. Participants rated the tool’s usability via the Website Analysis and Measurement Inventory (WAMMI, range 0–95) and their perceived 10-year and lifetime risk of cardiovascular disease (CVD) on a visual analog scale (range 0–10). Participants’ perceived risk, blood pressure, and body mass index were measured at baseline and three months after enrollment. Healthcare providers (HCP, n=5) completed the WAMMI to determine the usability and feasibility of incorporating the HerHeart tool into clinical practice.

Results

Adolescent participants and HCPs rated the tool’s usability highly on the WAMMI with a median of 79 (IQR 65, 84) and 76 (IQR 71, 84). At the baseline visit, participants’ median perceived 10-year risk of a heart attack was 1 (IQR 0, 3), and perceived lifetime risk was 2 (IQR 0, 4). Immediately after engaging with the tool, participants’ median perceived 10-year risk was 2 (IQR 1, 4.3), and perceived lifetime risk was 3 (IQR 1.8, 6). Thirty-one participants chose to set a behavior change goal, and 12 participants returned for follow-up. Clinical metrics were similar at the baseline and follow-up visits.

Conclusion

HerHeart is acceptable to young women and demonstrates potential for changing risk perception and improving health habits to reduce risk of CVD. Future research should focus on improving retention in studies to promote cardiovascular health within reproductive health clinics.

Introduction

Cardiovascular disease (CVD) is the leading cause of death for women, who have a lifetime risk of CVD of over 30%¹. CVD is also the leading cause of maternal morbidity and mortality in the United States². Many reproductive-aged women are unaware of their risk of CVD³ or how CVD risk relates to maternal morbidity and adverse pregnancy outcomes such as gestational diabetes, preeclampsia, and preterm birth^4,5. Increasing awareness of CVD risk and risk factor control among adolescent and young adult (AYA) women in the pre-pregnancy period is critical to improving the cardiovascular and maternal health of this and future generations^6,7.

A 2018 Presidential Advisory from the American Heart Association (AHA) and the American College of Obstetricians and Gynecologists (ACOG) called for gynecologists to incorporate CVD prevention into all well-woman visits⁸. Weight management, smoking cessation, physical activity assessment, nutritional counseling, and stress reduction were highlighted in the advisory as essential components of well woman care, which many women of reproductive age receive only from an obstetrician/gynecologist (OBGYN)⁹. The advisory also addressed the importance of beginning CVD screening soon after menarche, in advance of and in conjunction with contraceptive visits. Despite this, many women report low awareness of their CVD risk factors when attending outpatient OBGYN clinics, and OBGYN practitioners report time constraints as a major barrier to assessing CVD risk^10,11.

Studies with adult women show promise for integrating CVD risk assessment tools into pregnancy and post-partum visits¹². However, previous studies aiming to improve CVD awareness and preventive behaviors amongst AYA have typically focused on nonclinical populations, such as college students^13,14. Other studies targeting AYAs have attempted to improve only a few CVD risk factors at a time (such as smoking rates, nutrition behaviors, alcohol intake, physical activity levels, and/or obesity)¹⁵. Few interventions have targeted the full spectrum of cardiovascular health or focused on young women presenting to OBGYNs and reproductive health practices.

To address this gap, we employed an iterative, user-centered design process¹⁶ to adapt an adult CVD risk assessment tool, the Healthy Heart Score,¹⁷ for use with young women. Importantly, the Healthy Heart Score has been validated to predict the development of CVD risk factors before they occur in adult women specifically, and the risk of early CVD events in young adults^18,19. This new AYA-friendly web-based tool, HerHeart, assesses participants’ dietary patterns (intake of fruits and vegetables, sugar-sweetened beverages, nuts, whole grains, and red or processed meats), physical activity (hours of mild and moderate/vigorous activity per week), smoking status, alcohol intake, and body mass index (BMI). The tool then presents a relative CVD risk score and offers participants the opportunity to receive personalized feedback and set health behavior goals. We hypothesized that HerHeart, when delivered in conjunction with annual reproductive health and primary care visits, would be acceptable to young women and feasible to integrate into clinical practice.

Methods

We recruited 60 AYA women ages 13–21 years old from the waiting room of either an adolescent reproductive health practice (n=20) or general adolescent medicine practice (n=40). These two clinics, located in downtown Atlanta, Georgia, primarily serve patients covered by Medicaid. All participants completed informed assent/consent and the study was approved by the Institutional Review Board.

Study Procedures

Participants completed an initial survey consisting of demographic information and rated their perceived 10-year and lifetime risk of CVD on a visual analog scale (VAS) from 0 (“Never going to happen”) to 10 (“Definitely will happen”). Ten-year and lifetime CVD risk horizons were chosen as these are the standard time horizons used by the American Heart Association in adult risk assessment²⁰. Participants also rated how severe they believe a heart attack is on a VAS scale from 0 (“Like perfect health”) to 10 (“Like death/dying”). Participants were then instructed how to complete the HerHeart tool on a study tablet or their personal device.

Once participants completed the HerHeart assessment, the tool presented them with the option to receive personalized feedback about their CVD risk. Although the HerHeart tool generates a numerical 20-year risk of CVD based on the validated algorithm from which it was adapted, the HerHeart tool only displays an individual’s relative risk of CVD, as this is more germane to younger age groups who have a low short-term but high lifetime risk of CVD¹⁷. If participants chose to receive feedback, they were then prompted to choose two categories to improve from a list that was generated in response to their personal responses on the HerHeart assessment. Categories for improvement included Fruits and Veggies, Red and Processed Meats, Nuts and Seeds, Soda, Smoking, and Physical Activity. Alcohol was not included as a feedback option as it was assessed only for participants 21 years of age and older due to formative data suggesting that younger adolescents may not be truthful in their responses to this question¹⁶. BMI was not included as a feedback option due to formative data suggesting that young women found language around BMI stigmatizing and the clinical team’s concern that this may promote disordered eating¹⁶.

After completing the HerHeart tool, participants again rated their perceived 10-year and lifetime risk of CVD on the same VAS and reported their likelihood of recommending the tool to their friends and changing their health habits on a VAS from 0 (“Not likely at all”) to 10 (“Extremely likely”). Participants also rated HerHeart’s usability via the Website Analysis and Measurement Inventory (WAMMI, range 0–95) at their baseline visit. An above average score on the WAMMI is any rating greater than 47.5. Participants’ height, weight, and blood pressure (BP) were extracted from the medical record from the corresponding clinic visit. Height and weight were used to create body mass index (BMI) categories using centiles per the Centers for Disease Control and Prevention Recommendations.

Participants opting to continue into the 3-month intervention phase (n=55) received text messages from the study coordinator at 1- and 2-months post-enrollment reminding them to reflect on personal goals set at the conclusion of the HerHeart tool. These messages were concise and generalized to send to all participants and did not detail individual goals from the HerHeart tool. Participants were also reminded of the date and time of their 3-month follow-up with an option to reschedule if they were no longer available.

Three months post-enrollment, twelve participants returned for a final study visit where they had their height, weight, and BP measured by a clinical research coordinator. Participants were instructed to complete the HerHeart tool again; data was lost on two of the twelve participants who completed HerHeart at follow-up due to technical difficulties. After completing the HerHeart tool, participants again rated their perceived 10-year and lifetime risk of CVD on the same VAS used at baseline. Participants were compensated with a $25 gift card following their initial visit and a $35 gift card following their 3-month follow-up. Both the initial and three-month follow-up were completed in approximately 20 minutes each, of which 5 minutes or less was spent completing the actual HerHeart tool.

Healthcare providers (HCPs, n=5) from the same two clinical practices in Atlanta, Georgia were recruited by the research coordinator to understand their perspectives about incorporating the HerHeart tool into clinical practice. HCPs were surveyed via the WAMMI after completing the HerHeart tool. They were then asked to rate the usefulness of the HerHeart tool in their clinical practice on a scale from 0 (“Not at all useful”) to 10 (“Extremely useful”). HCPs were also asked to rate their likelihood of incorporating the HerHeart tool into their practice on a scale from 0 (“Not at all likely”) to 10 (“Extremely likely”). HCPs were compensated with a $25 gift card for their participation.

Statistical Analysis

Primary outcomes included feasibility of recruitment and retention, usability of the tool as measured by the WAMMI, and acceptability of the tool as measured by participant reactions via the survey. Secondary outcomes included reported health behaviors, clinical characteristics, and calculated 20-year CVD risk using participant responses to HerHeart. Median and inter-quartile range (IQR) were used for all descriptive statistics due to lack of normality in several statistics and small sample sizes. Given significant loss to follow up, statistical tests of group comparisons were not attempted to test any hypotheses. All descriptive statistics were calculated using R v.4.2 (Vienna, Austria).

Results

The median age of recruited participants at baseline was 17 (IQR 15, 18). Fifty participants self-identified as Black (83%); 7 (11.7%) participants self-identified as Some Other Race, 1 (1.7%) participant self-identified as White, 1 (1.7%) participant self-identified as an Asian/Pacific Islander, and 1 (1.7%) participant self-identified as a Native American or Alaskan Native. The median age of the twelve participants who returned for follow-up was also 17 (IQR 15, 17.8). Among these twelve participants, 10 (83.3%) self-identified as Black and two (16.7%) self-identified as some other race.

Primary Outcomes

A total of 90 patients were approached to gauge their interest in the study, of which 16 (17.8%) were not interested. After completing a screening questionnaire, 5 (5.6%) patients were ineligible to participate based on the study’s exclusion criteria (e.g., diagnosis of a mental health disorder, or pre-existing hypertension or diabetes). A total of 9 (10%) patients declined to participate due to a personal barrier to participation (such as transportation barriers or prior school or work commitments). Out of the 60 participants who were enrolled in the study, 5 (8.3%) withdrew and 43 (71.7%) were lost to follow-up, 14 from the reproductive health practice and 29 from the adolescent medicine practice. Therefore, 12 active participants remained throughout the entire 3-month duration of the study.

Participants (n=60) rated the HerHeart tool’s usability highly on the WAMMI with a median of 79 (IQR 65, 84). At baseline, participants’ median perceived 10-year risk of a heart attack was 2 (IQR 1, 4.3) after completing the HerHeart tool compared to 1 (IQR 0, 3) before completing HerHeart. Their perceived lifetime risk was 3 (IQR 1.8, 6) after completing HerHeart compared to 2 (IQR 0, 4) prior to completing HerHeart. After completing the HerHeart tool, participants’ perceived severity of a heart attack was 10 (IQR 8, 10) compared to 9 (IQR 7, 10) before completing HerHeart. Perceived 10-year risk and lifetime risk changes were sustained for the participants completing HerHeart again in 3 months.

Participants were likely to recommend HerHeart to their friends and report their intention to change their health habits immediately after completing the tool. Additionally, 31 participants (52%) chose to receive personalized feedback and set a health behavior change goal at baseline. At the three-month follow-up, five of the ten participants (50%) chose to receive feedback from the HerHeart tool again (Table 1).

Table 1.

Survey data from adolescent and young adult women completing the HerHeart cardiovascular risk assessment tool

	Immediately After Completing HerHeart Tool, N = 601	3-months After Completing HerHeart Tool, N = 121
Likelihood of Recommending to Friend (range, 0–10)	8 (7, 10)	6.5 (6, 9)
Likelihood of Behavior Change (range, 0–10)	8 (7, 10)	8 (5.5, 10)
Chose to Receive Personalized Feedback and Set Behavior Goals	31 (52%)	5 (50%)
Number of Participants Who Chose Each Area to Improve2
Physical Activity	20 (33%)	4 (27%)
Soda	13 (22%)	4 (27%)
Nuts and Seeds	9 (15%)	2 (13%)
Red & Processed Meats	8 (13%)	2 (13%)
Fruits & Veggies	6 (10%)	2 (13%)
Smoking	4 (7%)	1 (7%)

¹Median (IQR) or n (%)

²Percentages are based on respected sample size x the ability to choose 2 options

Data on behavior goals was extracted from the HerHeart Tool and missing for 2 of the 12 participants who returned for 3-month follow-up

Secondary Outcomes

Participant habits as captured in the HerHeart tool for the total sample at baseline are shown in Figure 1. The results are represented in the same visual style seen by individual participants completing HerHeart; however, Figure 1 displays the median values from all sixty participants at the baseline assessment. Median HerHeart results from baseline and follow-up are quantitatively shown in Supplemental Tables 1 and 2, respectively.

Figure 1.

Legend: Data visualization is representative of the style of personalized feedback given to individual participants although depicted values represent the median for the population. The figure was created using Figma to match the style of what participant results look like in the HerHeart tool.

At baseline, participants had a median BMI of 24 kg/m² (IQR 21, 28), systolic BP of 118 mmHg (IQR 112, 121), and diastolic BP of 67 mmHg (IQR 63, 71). At the three-month follow-up, the 12 participants who returned had a median BMI of 24 kg/m² (IQR 23, 33), systolic BP of 123 mmHg (IQR 112, 134), and diastolic BP of 67 mmHg (IQR 63, 73) (Table 2). Twenty-year risk of a CVD event per the validated algorithm was 0.10% (IQR 0.07%, 0.13%) for the total sample and 0.08% (IQR 0.06%, 0.10%) for the 10 participants who completed HerHeart at the three-month follow-up. The 10 participants who completed HerHeart at the three-month follow-up reported a median of 1.4 (IQR 0.8, 2) hours of moderate physical activity per week, compared to a median of 0.8 (IQR 0.6, 2) hours for all participants at baseline. Furthermore, these 10 participants reported a median of 0.3 (IQR 0.1, 0.5) weekly servings of processed meats and 2 (IQR 1, 2) daily servings of vegetables at the three-month follow-up, compared to a median of 0.5 (IQR 0.3, 1) weekly servings and 1 (IQR 0.9, 2) daily servings, respectively, at baseline. Data were similar when the sample was restricted to only the 12 participants that completed the baseline and follow-up visits (Supplemental Table 3).

Table 2.

Clinical characteristics of adolescent and young adult women completing the cardiovascular HerHeart risk assessment tool

	Immediately After Completing HerHeart Tool, N = 601	3-months After Completing HerHeart Tool, N = 121
Clinical Characteristics
BMI	24 (21, 28)	25 (23, 35)
Underweight	0 (0%)	0 (0%)
Healthy Weight	25 (49%)	5 (42%)
Overweight	18 (35%)	2 (16%)
Obesity	8 (16%)	5 (42%)
Systolic Blood Pressure	118 (112, 121)	117 (108, 133)
Diastolic Blood Pressure	67 (63, 71)	67 (62, 72)
Smoking Status
Never	46 (77%)	9 (90%)
Past	4 (6%)	1 (10%)
Current	10 (17%)	0 (0%)
Missing	0	2

¹Median (IQR) or n (%)

²Percentages are based on respected sample size x the ability to choose 2 options

Data on smoking and behavior goals was extrapolated from the HerHeart Tool and missing for 2 of the 12 participants who returned for 3-month follow-up

Healthcare Provider Survey Responses

The five HCPs rated the HerHeart tool’s usability highly on the WAMMI with a median of 76 (IQR 71, 84). HCPs found the HerHeart tool useful with a median of 9 (IQR 9, 10) and were likely to incorporate it into their practice also with a median of 9 (IQR 6, 9).

Discussion

The developmentally appropriate web-based cardiovascular risk prediction tool, HerHeart, shows promise for increasing the awareness of CVD risk among young women presenting to reproductive care and adolescent health visits. Participants in this study found the HerHeart tool to be user-friendly, reported being likely to recommend it to friends, and reported intention to change their health habits after engaging with the tool. Over half chose to receive personalized feedback and set a specific heath goal. Most participants had a normal BMI and blood pressure at baseline, and these measures were essentially unchanged at follow-up.

Improving CVD risk awareness with the HerHeart tool’s personalized feedback is an important first step toward improving young women’s health habits and ultimately promoting cardiovascular health and reducing maternal morbidity and mortality^21,22. Importantly, while AYA women in general tend to underestimate both their 10-year risk and lifetime risk of CVD,²³ most participants in this study overestimated their 10-year risk yet still underestimated their lifetime risk. Future refinements to the tool may be necessary to further improve the accuracy of short and long-term risk perception, which can be challenging in younger individuals^3,24–26. Whether accurate CVD risk perception improves health behaviors over time and actually reduces maternal morbidity in young women is an important area for future study.

HerHeart could assist reproductive healthcare providers as they counsel young women to adopt heart healthy behaviors to reduce their medium-term risk of maternal morbidity and their lifetime risk of cardiovascular disease. In accordance with the 2018 Presidential Advisory from AHA and ACOG, a tool such as HerHeart could be used at annual well-woman visits and contraceptive counseling visits. While healthy young women are unlikely to be particularly concerned about their cardiovascular health at these early adolescent and young adult visits²⁷, incorporating CVD risk early could set their expectation that this will occur annually as they age. The HerHeart tool could be particularly helpful for special populations with increased risk for CVD such as young women with early or late menarche, irregular menses, or polycystic ovarian syndrome (PCOS)^28–31. HerHeart would add an additional benefit to young women who have a family history of premature myocardial infarction since these women have an increased prevalence of CVD risk factors³².

Prior research has found that CVD risk trajectories develop during early adulthood and may occur even with a normal BMI³³, especially in patients with PCOS³⁴. In contrast, studies have shown that adolescents who maintain physical activity and practice healthy dietary habits throughout young adulthood have a lower risk of CVD^35,36. Other studies have demonstrated that intervention in the early stages of adulthood can promote healthy habits that ultimately reduce future CVD risk^33,37. Our study adds to this body of evidence by targeting CVD risk awareness in adolescent and young adult women presenting for reproductive health care specifically.

Additional research is also needed to determine the optimal constellation of behavior change techniques beyond simply raising risk awareness. Some studies have found additional behavior change techniques to be effective in young adults, such as practicing mindfulness to promote lifestyle self-management, fostering healthy habit formation and stressing the salience of consequences^38,39. Others have found that in individuals with a lower socioeconomic position such as those in this study, focusing on fewer behavior change techniques might be more effective for lasting behavior change^40,41. Peer coaches, social support, and social accountability may also encourage young women’s engagement and retention in behavior change trials^42,43. Finally, consideration of the family, neighborhood, and school environment may be necessary as many adolescents do not have complete control over their dietary choices and might be more affected by environmental factors.

The current study has several limitations. Although the study met our recruitment goal of 60 participants, only 12 of the 60 participants returned for follow-up despite monetary incentives and monthly text message reminders. Due to our limited sample size, results should be interpreted with caution and are not generalizable beyond our local sample. Multiple studies have demonstrated challenges in recruitment and retention of young adults participating in nutrition, physical activity and/or obesity intervention studies, with less than one-third of young adults who initially express interest in an intervention ultimately providing consent and participating in a study⁴⁴. For our study, participants may have experienced personal barriers (such as organizing transportation and scheduling time off from work or school) that limited their ability to return to the clinic for follow-up. Our study design sought to confront these logistical concerns by recruiting participants who were already present in the clinic for a visit and scheduling the three-month visit on the same day as a future clinic visit, such as for contraceptive follow-up; however, most participants did not have an appointment scheduled within the study timeframe of three months. In a future iteration, scheduling follow-up visits in accordance with patient clinic appointments (e.g., 6 months or 1 year) may improve participant retention.

Prior studies have successfully increased young adult engagement in mobile health (mHealth) trials through consistent human outreach, study reminders, and assurance that participant interpretation of the clinical trial is accurate⁴⁵. In the current study, more specific and personalized monthly text message reminders sent by the research coordinator pertaining to participants’ unique goals could have been more effective in engaging participants. For example, a participant who selected Fruits & Veggies as a category to improve in could receive messages targeted to this specific goal. Additionally, taking more time to explain the requirements of the study to participants could have resulted in a better understanding of the purpose of the clinical trial and their role⁴⁵. Offering virtual visits at times convenient for participants could reduce participant burden resulting from travel expenses, prior schedule commitments, and unforeseen circumstances that may prevent them from returning in-person. Future studies should partner with AYAs to design ways to engage and motivate adolescents to foster retention and consider adjusting recruitment goals to account for challenges in retention of participants⁴⁶.

Despite these challenges, the HerHeart tool was rated highly by adolescents and shows promise for improving risk perception and motivating changes in lifestyle and dietary habits among young adult women presenting for reproductive health care and annual wellness visits. Tools such as HerHeart could assist reproductive healthcare and adolescent providers in busy clinical practices to counsel young women about heart healthy behaviors, which should be rigorously evaluated through implementation studies. Given the short-term risk of maternal morbidity and mortality and the lifetime risk of cardiovascular disease for young women with CVD risk factors, effective and feasible interventions for this population are urgently needed.

Abbreviations

CVD

Cardiovascular Disease

AYA

Adolescent and Young Adult

HCP

Healthcare Provider

WAMMI

Website Analysis and Measurement Inventory

BMI

Body Mass Index

VAS

Visual Analog Scale

BP

Blood Pressure

IQR

Inter-Quartile Range

mHealth

Mobile Health

OBGYN

Obstetrician and Gynecologist

ACOG

American College of Obstetricians and Gynecologists

AHA

American Heart Association

PCOS

Polycystic Ovarian Syndrome