Perceived versus Actual Risk of Cardiovascular Disease in College Students

Background: College wellness interventions have potential to reduce the growing burden of cardiovascular disease (CVD) by guiding students to establish behaviors for lifetime CVD risk reduction. Yet, some students may not see CVD prevention as a relevant issue during young adulthood. Purpose: We examined perceived risk of CVD amongst 148 college students and compared perceived risk to demographic characteristics and traditional risk markers. Methods: Survey data on perceived CVD risk and student characteristics were linked to biochemical lab values from a cholesterol screening program. Results: Only 39.2% of students had been previously screened for dyslipidemia. Mean perceived CVD risk was highest amongst students with age ≥21 years, BMI ≥30, a family history of CVD, and a fair or poor rating of overall health. Perceived risk was not higher amongst students with total cholesterol values outside of the normal range. Discussion: While some students were aware they may possess an elevated CVD risk, we identified others who were unaware of a potential risk. Translation to Health Education Practice: These variations in perceived risk indicate that individualized intervention approaches may be necessary to engage college students in behaviors for lifelong CVD prevention.

Cardiovascular disease (CVD) affects nearly half of all U.S. adults and is a major contributor to morbidity, mortality, and increased health care costs.[1] Estimates from 2015 data show the direct and indirect costs attributable to CVD and stroke to be around $351.2 billion, and this number is expected to increase in the coming years.[1] As much as 80% of CVD morbidity and mortality is considered preventable by addressing modifiable risk factors such as physical activity and diet. Thus, health education interventions which emphasize engagement in these behaviors over the life course can play an important role in reducing the population burden of CVD.[2],[3] Because CVD does not typically manifest until middle or older age, interventions often target these age groups. There is ample opportunity to reduce CVD risk in adult populations: less than one quarter of U.S. adults report meeting guidelines for physical activity, more than one-third are obese, and an estimated 95 million have elevated levels of total cholesterol.[1] Though CVD prevention is not typically an area of concern among younger populations,[4] strong evidence exists that the pathogenesis of CVD can start as early as childhood.[5]–[11] By targeting younger populations with interventions that focus on implementing healthy behaviors for long-term CVD prevention, the burden of disease in middle and older age could later be reduced. As such, a wide variety of intervention programs are being implemented which focus on engaging children and young adults in preventive behaviors that will reduce their lifetime risk of CVD.[12],[13]

The college period, typically encompassing the ages 18–24, is an ideal time to implement wellness interventions that seek to establish and reinforce patterns of behavior that can have a lasting impact on disease risk later in life. This time can be quite powerful as many young adults are transitioning to independent living for the first time, and acquiring the freedom to make independent decisions about their personal health. While some college students may be able to independently establish health-promoting behaviors, many others could benefit from wellness programming that motivates and supports them in establishing healthy patterns of behavior. This is particularly important for CVD prevention: many young adults already possess behaviors that could contribute to the development of CVD later in life. Specifically, findings from the National College Health Assessment indicate that 90% of college students eat fewer than five servings of fruits and vegetables per day, and less than 30% participate in recommended amounts of physical activity.[13]–[15] Research on this population has also documented that 6.8 to 9.0% of college students meet the classifications for metabolic syndrome, and between 27.0 to 77.0% of students have at least one component of metabolic syndrome.[5]–[7],[16]–[18] CVD risk in young adults is expected to increase in coming years: studies using time trend data indicate that body weight has been steadily increasing and aerobic fitness steadily decreasing in populations of college students over time.[19] The college student population has also become increasingly diverse in recent decades.[20] Given research showing that CVD risk is elevated in ethnically diverse populations,[21],[22] college wellness interventions have a critical role to play in early CVD prevention amongst high risk groups. College wellness interventions that address the initiation and maintenance of heart healthy behaviors have great potential to reduce the future burden of CVD across demographic subgroups.

Many colleges and universities are uniquely positioned with the resources to assist students in establishing health-promoting behaviors that will reduce their risk of CVD over the life course. Colleges have access to recreation and fitness facilities, on-site faculty and staff expertise, and the unique ability to engage students in wellness courses of varying duration and intensity. Many universities have established employee wellness programs run by partner health care organizations, providing access to the medical resources necessary for conducting biometric screenings in students. Great potential exists to increase college students' understanding of the prevalence and prevention of CVD: many students enter college with limited knowledge of CVD and limited awareness of their own risk of developing CVD over the life course.[23],[24] Despite prime opportunities to implement behavior change interventions in the university setting, it may be challenging to motivate college students who have limited awareness of their likelihood of developing CVD. A number of health behavior models emphasize that one's perceived susceptibility for acquiring a disease can greatly influence the motivation to adopt preventive behaviors.[25]–[27] College students may not consider CVD prevention as immediately relevant because they do not perceive themselves to be at risk. This, in turn, could impact their motivation to adopt and prioritize behaviors for CVD prevention.

Understanding patterns of perceived CVD risk amongst college students could assist educators and interventionists in the design of college wellness programs that emphasize behavior change for CVD prevention. In order to contribute data on the role that perceived risk of CVD could play in college students' motivation to engage in heart healthy behaviors, we measured levels of perceived CVD risk and compared this to traditional markers of risk. Specifically, we used data from a college wellness course to explore students' perceptions of their CVD risk alongside socio-demographics, BMI, family history, overall health, physical activity, and biochemical risk markers gathered from a lipid profile screen. We sought to identify subgroups of college students who may be overly optimistic about their risk for developing CVD, and subgroups of students who are already aware that they possess an elevated risk of CVD.

The study was conducted at an undergraduate liberal arts university located in a suburban area of the Southeastern U.S. (undergraduate university enrollment in year of survey = 2,797). Study participants were recruited from a 4-credit semester-long "Wellness Concepts" course that covers multiple dimensions of wellness with an emphasis on increasing students' knowledge of and competence in behaviors for lifetime chronic disease risk reduction. The curriculum includes extensive instruction in the primary prevention of CVD with an emphasis on physical activity and nutrition. The semester-long course required students to attend two days of lecture and participate in two, fifty minutes sessions of guided physical activity each week. A lipid profile screen is offered in the second week of the course as a technique for increasing students' awareness about their personal cholesterol and the components of biochemical lab testing. Because the course fulfills a requirement for graduation, nearly all students attending the university enroll in the course at some point during their college career. Each "Wellness Concepts" course section is comprised of students from a range of undergraduate majors and class years.

All study participants consented to participate in a brief survey and biochemical blood draw conducted as part of the course's cholesterol screening program. The survey consent forms, and study protocol were approved by the University's Institutional Review Board (Furman University IRB Study number 120417). The survey was administered using pen and paper during class time by faculty in eleven sections of the "Wellness Concepts" course. All surveys were completed in the first week of classes (January 14–18, 2018 and September 28–30, 2018), before any educational instruction on CVD had begun, and prior to the biochemical blood draw. The survey was comprised of questions assessing students' perceived levels of lifetime CVD risk, along with socio-demographics (age, gender, race/ethnicity), physical activity, family history of heart disease, and self-rated general health. Perceived risk of CVD was measured using the 7-item dread risk subscale from the perception of risk of heart disease scale (PRHDS).[28] The dread risk scale included questions regarding the likelihood of developing CVD over the life course (see Figure 1 for the 7 questions comprising the scale). Responses to questions were categorized using a 4-point Likert Scale (strongly disagree to strongly agree). Scores were summed to obtain a total "dread risk score," and a higher score indicated a higher perceived risk of CVD. The reliability of the dread risk scale has been demonstrated previously: the Cronbach's alpha score for internal consistency was 0.79, and test-retest reliability was 0.76.[28] Physical activity was assessed via a modified version of the International Physical Activity Questionnaire (IPAQ).[29],[30] The IPAQ assesses both type and frequency of physical activity through self-report, and has been shown to be reliable in previous studies, with interclass correlation coefficients (ICC's) ranging between 0.71 and 0.89 in samples of college students.[29],[30] IPAQ data was categorized according to whether or not U.S. DHHS guidelines for physical activity were met (75 minutes of vigorous or 150 minutes of moderate exercise per week).[31] Body Mass Index (BMI) was calculated using self-reported height and weight, and data was categorized into standard groups. Self-rated general health was measured on a 5-point Likert scale ("excellent", "very good", "fair", or "poor").[32],[33] Students were asked to report the last time their cholesterol was checked and to recall the number of close family relatives that they had who had experienced heart disease (defined as hypertension, high cholesterol, a heart attack, stroke, or heart failure).

Graph: Figure 1. PRHHS Dread Risk Scale. Total scores were calculated by summing responses: Strongly Disagree = 1, Disagree = 2, Agree = 3, Strongly Agree = 4.

Fasting venous blood samples were drawn and lipid profiles were analyzed through an existing collaboration with a large health care system. Samples were analyzed for total cholesterol (TC), triglycerides (TRIG), low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C). Data were then stripped of all identifiers and entered into a database for statistical analysis. Results were then classified as falling within or outside of the normal range, in accordance with clinically relevant reference ranges (TC ≥200, HDL<50 (females) or <40 (males), LDL ≥200, or TRIG ≥150).[34] All labs were reviewed by clinical staff, and abnormal values were flagged for follow-up.

De-identified survey data were linked by unique student identifier to biochemical lab data, and PRHDS dread risk scores were calculated for each individual. Descriptive statistics were calculated to provide information regarding demographics, behavior, perceived CVD risk, perceived general health, and biochemical markers of health. Mean PRHDS risk were calculated by each categorized value, and t-tests, and one-way analysis of variance (ANOVA) models were used to determine whether dread risk scores were statistically different across categories. Finally, dread risk scores were divided into tertiles, and the percent of lab values outside of the normal range were examined for those in the lowest tertile (score of <11) compared to the rest of the sample (score ≥11). All statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC).

Of the 275 students invited to participate, 243 completed the survey. Of these, 148 also participated in the blood draw and were included in the current study. Characteristics of the study population are shown in Table 1. The majority (84.3%) of students were white, and 68.2% of the sample was female. The majority (70.8%) of students were in the normal weight BMI category, 17.0% of students were in the overweight, and 8.8% were in the obese BMI category. Only 39.2% had been previously screened for dyslipidemia, 13.5% rated their overall health as poor or fair, and 32.4% did not meet U.S. DHHS guidelines for vigorous or moderate physical activity.

Table 1. Characteristics of study participants (n = 148).

1 *U.S. DHHS Guidelines for physical activity: 75 minutes of vigorous and/or 150 minutes of moderate exercise per week

Table 2 shows biochemical values from lipid screening tests. Of students in the study, 8.8%, 24.3%, 22.3%, and 6.1% had a lab value outside of the normal range for TC, HDL-C, LDL-C, and TRIG, respectively. When biochemical data were examined in aggregate, 39.2% had 1 or more biochemical value outside of the normal range, and 14.1% of students had 2 or more biochemical values outside of the normal range. Figure 2 shows Mean PRHDS Dread Risk Scores (perceived CVD risk) stratified by participant characteristics. Perceived CVD risk was significantly higher amongst students who were older (≥21 years old), who reported a family history of heart disease, and who rated their general health as fair or poor. Perceived risk was also highest among students in the obese category (BMI ≥30 kg/m2), yet differences in perceived risk between the four BMI categories were marginally significant (F = 2.59, p =.055). Perceived CVD risk did not differ by gender, race, and/or whether or not a student met DHHS guidelines for moderate or vigorous physical activity. There were no differences in perceived CVD risk scores between students with TC values outside of the normal range and students with TC values within the normal range.

Table 2. Biochemical values of study participants (n = 148).

PHOTO (COLOR): Figure 2. Perceived risk (mean dread risk) score* by participant characteristics (n = 148). *Higher Dread Risk Score = Higher Perceived CV Risk

When we examined students most optimistic about their CVD risk (students with dread risk scores in the lowest tertile of the sample), we found that 30.2% of this group had one or more lab values outside of the normal range and 12.8% had two or more lab values outside of the normal range. Figure 3 shows the percentage of students with abnormal screening tests by BMI category (normal and underweight vs. overweight and obese). While the percentage of students with screening tests outside of the normal range was higher among students with BMI ≥25 kg/m2, a substantial portion (32.9%) of students with BMI <25 kg/m2 still had at least one biochemical lab value outside of the normal range.

Graph: Figure 3. Percent of students with abnormal* screening tests by BMI status. *Screening tests outside of normal range defined as: TC > 200, LDL ≥ 100, TRIG ≥ 150, and HDL < 50 females.

The college environment is an ideal setting to implement wellness interventions to assist students in establishing behaviors to reduce their risk of CVD over the life course. However, some college students may be less receptive to CVD prevention messaging because they do not perceive diseases that develop over a long period of time to be relevant during young adulthood. Using data from students enrolled in a college wellness course, we examined perceived risk of CVD, and assessed relationships between perceived risk, demographic and behavioral variables, and known CVD risk factors. Only 39.2% of students had been previously screened for dyslipidemia. Perceived CVD risk was highest amongst students who held traditional CVD risk markers such as high BMI and knowledge of a family history of heart disease, as well as amongst those who rated their overall health as fair or poor. Perceived risk did not differ by gender, but was lower among students who were younger (<21 years old). Finally, our data revealed subgroups of college students who rated their perceived risk as low despite possessing abnormal lab values. Such students are a key population to consider for college wellness interventions that seek to expand students' awareness of CVD risk, and motivate them to engage in behaviors for lifelong CVD prevention.

Data from this study fill a gap on the concordance of perceived CVD risk with established markers of risk in populations of young adults. Studies conducted in populations of middle aged and older adults have shown that perceptions of CVD risk do not accurately align with actual markers of risk,[35]–[38] yet there is limited data examining this alignment in populations of younger adults. PRHDS dread risk scores among the college students in our study were markedly lower than PRHDS scores reported in studies of older populations.[39] Previous research has also documented that young adults consistently perceive their risk of CVD to be low, and tend to be optimistic regarding lifetime levels of CVD risk.[40]–[43] The current study builds upon existing research in young adult populations: our results provide evidence that some students possess an elevated risk of developing CVD despite perceiving their risk to be low. With the prevalence of CVD nearing 50% among adults, it is likely that a substantial portion of college students will go on to develop CVD at some point over their life course.[1] Thus, this "optimistic bias" that can be so prevalent amongst young adults can create a challenge for practitioners working to engage college students in establishing lifetime CVD risk reduction behaviors. Students who do not feel vulnerable to developing CVD may be resistant to messages regarding the initiation of and/or sustained engagement with health promoting behaviors.[44],[45]

The National Cholesterol Education Program (NCEP) recommends that routine testing of cholesterol should begin in young adulthood as a method for early identification of CVD risk.[46] Despite this recommendation, nationally representative data show that fewer than 50% of young adults report having ever been screened for dyslipidemia.[47] In our sample, only 39% of students reported that they had previously had their cholesterol tested. The reasons for low screening rates in young adults are varied: these populations are less likely to be enrolled in workplace wellness programs, and are less likely than middle and older-aged adults to receive regular preventive care.[48],[49] Also, despite the NCEP recommendations, screening for dyslipidemia is not standard practice amongst most U.S. health care providers: the U.S. Preventive Services Task Force (USPSTF) recommends that cholesterol screenings should only be conducted among younger patients (men between the ages of 20–35 and women between the ages of 20–45) who are believed to be at high risk of CVD.[50] With population-based data showing 50 to 55% of young adults in the U.S. have at least one CVD risk factor, these existing gaps in cholesterol screenings represent missed opportunities to raise awareness around CVD risk in younger populations.[47] Integrating cholesterol screenings into college wellness interventions would provide students with individualized health information that could increase their motivation to adopt and prioritize preventive behaviors.

We identified subgroups of students who perceived their CVD risk to be higher than their peers, including students ≥21 years, students who were aware of a family member with heart disease, and students whose BMI was in the obese category (≥30 kg/m2). Perceived CVD risk also increased as students' self-rating of their overall health decreased: the highest levels of perceived risk were found among those students who rated their general health as "fair" or "poor". Our finding that a family history of heart disease increases one's perceived risk of developing heart disease was expected, and in accordance with previous research.[39] The higher levels of perceived risk found amongst students with lower self-rated general health may also be a useful finding. The global measure of self-rated health is a simple measure that provides information both on one's physical health state, as well as on one's health behaviors.[51] Because BMI aligns with self-rated health in young adults, this simple general health question could be useful as a way to globally screen for students who may require closer follow-up.[52] The self-rated health question could also be used as a way to engage students in a larger conversation around their readiness to adopt healthy behavior patterns.

We identified subgroups of students who rated their CVD risk as low despite possessing lab values outside of the normal range. These results illustrate the existence of college students who may be unaware they possess an underlying risk for developing CVD. Our data also showed that substantial portions of students whose BMI values fell at or below the "normal" range (BMI ≤25) still possessed biochemical lab values outside the range of normal. While the presence of 1 or 2 abnormal lab values does not ensure that a student will later develop heart disease, students who are unaware of their biochemical risk markers may inaccurately be using BMI as a singular marker of CVD risk. There has been recent attention given to the importance of extending assessment of cardio-metabolic risk beyond BMI: when used in isolation, BMI is not useful in identifying individuals who may be metabolically obese but of normal weight, and may contribute to the underestimation of CVD risk.[53],[54] Adding biometric screening programs to college wellness interventions would encourage a broader assessment of cardiovascular health, while also expanding opportunities to increase students' knowledge of the components of cardiovascular health and metabolic syndrome. Importantly, such screening programs would allow for the identification of elevated biomarkers in students who do not meet the BMI classifications of overweight or obese.

It is estimated that between 70 and 90 percent of colleges and universities offer some sort of wellness programming, ranging from short educational outreach sessions to intensive, semester long wellness courses.[55],[56] While wellness programs can cover a wide range of topics, most place a strong emphasis on both increasing knowledge about how CVD develops, and establishing behaviors for lifetime risk reduction such as physical activity and nutrition.[55],[57] A 2015 review of 41 college wellness interventions revealed that the greatest impacts on health outcomes occur when interventions include frequent face-to-face interactions with faculty or staff. This review also emphasized that the most successful interventions allowed students the opportunity to receive feedback on their progress toward health goals rather than relying on information-only approaches such as stand-alone lectures or brochures.[12] Our study extended upon this work, illustrating that some college students may be overly optimistic about their lifetime risk of CVD. Thus, the effectiveness of behavior change interventions could be increased by placing greater emphasis on expanding students' awareness of their personal CVD risk. Strategies to increase knowledge about the development of CVD could be paired with personalized feedback sessions on individual CVD risk to allow for face-to-face interactions. In this model, students could also be given opportunities to reflect upon personal strategies for behavior change, and to increase their self-efficacy to engage in such behaviors.[58],[59] For example, results from biometric screening programs or food frequency questionnaires could be paired with mentored sessions in which students outline their goals for CVD risk reduction. Instructors in wellness courses could also incorporate assignments that ask students to interview family members about CVD, and to reflect upon the implications of such for their individual patterns of behavior. Finally, researchers should consider that wellness programs may need to be intensive in both contact hours and duration to be effective. For example, in a 2015 study conducted by Melnyk and colleagues,[60] researchers met with students over the course of 3 short visits to discuss personalized blood pressure and cholesterol reports and distribute informational pamphlets on heart disease, diet, and physical activity. This individualized approach was effective in improving college freshmen's knowledge of heart disease. However, the authors noted that a longer and more in-depth intervention program may be needed to affect behavior change. Wellness courses which span the course of a semester allow opportunities to mitigate this limitation.

This study has many strengths: the availability of data on both biochemical values and perceived risk allowed for the unique ability to examine perceived versus established markers of risk. Previous studies have reported the prevalence of dyslipidemia and CVD risk factors amongst college students; however, to our knowledge, this study is the first to examine the concordance of these measures with an assessment of how students perceive their CVD risk. The wellness course used as a sampling frame in this study fulfills a general education requirement, and nearly every student at the university enrolls in the course before graduation. As a result, a range of class years and majors is represented, reducing potential for bias that can occur in studies of health-focused volunteers. Despite obtaining a high response rate for the survey (88.3%), our study sample was limited to the 148 students who both completed the survey and the biochemical blood draw (53.8% overall response rate). Despite this, our post-hoc analyses showed no differences in levels of perceived risk or BMI between the students included in our sample and those who completed only the survey. Further, while the study sample was limited to the course sections of instructors who agreed to participate, the distribution of biochemical lab values in our sample was similar to the distribution of lab values amongst all students in the study year and in previous years. While we are aware that data on blood pressure and more detailed measurements of body composition would have allowed for a more in-depth analysis of CVD risk, the current study was limited to data that was routinely collected as part of the wellness curriculum. Finally, while the small sample size only allowed for these analyses to be considered exploratory, the trends identified in this study are useful for generating hypotheses that can be examined in larger samples. Future studies conducted in larger populations would allow for more detailed subgroup stratification, including an examination of differences by socio-demographic factors.

The current study was conducted at a liberal arts college that draws students primarily from the Southeastern United States, and the participants in our sample were majority white and female. This does limit the ability to generalize our findings to larger universities with more socio-demographically and geographically diverse student populations. However, it is important to note that our results are also in line with similar studies conducted in diverse populations. For example, in a population of students attending a historically Black university, Sarpong and colleagues found students held low levels of awareness about individual markers of CVD risk (cholesterol, BP, and glucose values).[61] In another study, Holland and colleagues demonstrated the feasibility and efficacy of an intervention aimed to increase African-American students' awareness of CVD risk and assist them in strategies for risk reduction.[62] These similarities across studies highlight the potential for college wellness programs to play a key role in improving students' awareness of CVD risk across varying populations.

Interventions designed to engage young adults in behavior modification for lifetime CVD risk reduction have great potential to reduce the growing population burden of heart disease. This study contributes data on factors associated with college students' perceived susceptibility for developing CVD, and documents the extent to which students' perceptions of CVD risk align with traditional risk markers. Our data reveal that distinct subgroups exist among a sample of students participating in a wellness curriculum that emphasizes lifetime CVD risk reduction. Some students perceived their risk to be low despite having biochemical markers that are outside of the normal range. Other students perceived their CVD risk to be high – likely because they were aware that they held traditional markers of risk such as high BMI and/or family history. These variations in perceived risk suggest that university wellness interventions that use a one-size-fit-all approach may not effectively engage all students in behavior change for lifetime CVD risk reduction. For example, those students who perceive their CVD risk to be low may not be motivated to engage in heart healthy behaviors. Previous researchers have stressed that wellness curricula should be structured so that the goal of increasing knowledge of CVD risk is combined with proven strategies for behavior change (ex., enhancing self-efficacy, motivation, and self-reflection).[63] Thus, Health Educators and Certified Health Education Specialists who design and evaluate college wellness interventions need to consider strategies that address the varying levels of motivation that can exist within a group of students. Interventions should include the use of individualized strategies that can move each student along their own continuum of behavior change. For example, biometric assessments paired with face-to-face instructor follow-up and/or reflection assignments could increase student motivation to prioritize behaviors for lifetime CVD risk reduction. Further research is needed to evaluate the impact of these and other individualized strategies to increase student motivation for adopting and prioritizing heart healthy behaviors.

No potential conflict of interest was reported by the authors.