Examining the Impact of College Socioeconomic Status on Lipid Parameters: A Study in Targeted Pediatric Lipid Screening

Background and Objectives The Southern region of the United States is home to substantial populations with obesity, dyslipidemia, and hypertriglyceridemia, while also housing a large percentage of America’s minority, rural, and low socioeconomic status (SES) peoples. Modification of adult cardiovascular disease (CVD) may be improved by investigating associations(s) between youth demographic variables and lipid values. Our objective was to investigate lipid parameter associations with college-age socioeconomic status, which may improve age-specific screening algorithms for management or prevention of adult-onset CVD. Methods Using an Analysis of Variance test and a general linear model, associations between gender, race/ethnicity, SES, and athletic participation on lipid parameters (VLDL-C, LDL-C, TG, TC, and HDL-C) were analyzed in 4423 private liberal arts college students enrolled in freshman-level wellness courses at Furman University in Greenville, SC. Comparative data were collected from an age-matched sample (National Health and Nutrition Examination Survey: NHANES 2003-2016). Our main outcomes were statistically significant relationships between any lipid values (TC, HDL, LDL, TG) and any demographic variables (gender, SES, ethnicity, athlete status). Results Males demonstrated lower TC and LDL-C, and higher HDL-C values. HDL-C was highest in athletes. African-American students demonstrated healthier VLDL-C, TG, and HDL-C values. With similar distributions, the age-matched NHANES comparison group showed unhealthier values in nearly all categories. Conclusions College students may have better lipid health than the general population. African-Americans may have seemingly healthier lipid values than age-matched individuals independent of athletic or college enrollment which has already been demonstrated in other studies. Future research should include SES relationships in lipid screening paradigms along with other appropriate risk factors for cardiovascular disease. Based on our comparative data, pediatric health providers and

3 researchers may consider education as a potential protective factor against poor lipid health when considering lipid screening protocols for students.

Background The Center for Disease Control and Prevention lists the Southern region of the United
States as a leader in cardiovascular disease (CVD) mortality 1 . The South is home to significant populations living with obesity, dyslipidemia, and hypertriglyceridemia, and houses many individuals from minority, rural, and low socioeconomic status (SES), all of which are independent risk factors for adult CVD 2 . Unfortunately, individuals with the poorest lifestyle choices are less likely to have any prior cholesterol or blood pressure testing performed 3 . An increased prevalence of unhealthy lifestyles in both low SES and minority groups 3 and the population at large 4 has continued to predispose adults toward poor cholesterol health and other comorbidities that lead to CVD.
Research in CVD demonstrates that atherosclerosis begins with a long asymptomatic phase beginning in early adolescence or even childhood 5  for "Insufficient Evidence" 6 . Nonetheless, many pediatric clinicians still implement lipid screenings, subsequent education, and appropriate treatments for hyperlipidemia, if for no other reason than to identify patients with inherited dyslipidemia (which can occur in as many as 1 in 250 children who, if untreated, would otherwise maintain unhealthy LDL levels and may develop coronary artery disease sooner 7 ) in accordance with the American Academy of Pediatrics (AAP) 8 , the American Heart Association (AHA), the National Heart, Lung, and Blood Institute (NHLBI), and the National Lipid Association 9 . In addition, medical communities who screen their teenage patients assert that late adolescence represents an ideal age during which lipid screening can identify those at risk for atherosclerosis in adulthood from genetic and modifiable risk factors 9,10 . Lipid screening research by the NHLBI and others demonstrate that total cholesterol, triglycerides, body mass index, and systolic blood pressure measured at ages 15-18 have accurately identified individuals at risk for adult CVD 7,10 . Therefore, screening for poor lipid health and other CVD risk factors (including low SES and minority and ethnic heritage) during late youth or early college years may help diagnose premorbid cardiovascular disease and thus help prevent future vascular complications and/or death 10,11 . However, more information on the relationship between lipid values, SES, race/ethnicity, and lifestyle in youths/youths is needed to help resolve these disparate recommendations while also screening for early onset lipid disease in pediatric populations.
The purpose of this study was to investigate CVD risk factor associations in late youths/youths to provide additional information that may potentially improve age-specific screening algorithms for prevention of adult CVD. This study is innovative because to our knowledge, no studies have been performed that compare SES/ethnicity and CVD risk factors among undergraduate college students enrolled in a private, liberal arts university as a part of a wellness graduation requirement program, while also comparing results to an age-matched comparison group from a nationally representative database, the National Health and Nutrition Examination Survey (NHANES). All analyses for the Furman cohorts were carried out using SAS Enterprise Guide version 9.4 (SAS Institute, Cary, NC). The values of VLDL-C, TC, TG, LDL-C, and HDL-C were analyzed and presented as mean and 95% confidence intervals. An ANOVA test was used to determine independent association(s) of race, financial need status and gender on lipid parameters. Adjusted mean parameter values were obtained by using a general linear model; Tukey's studentized adjustment was made for multiple pair-wise comparisons. Pvalues <0.05 were considered statistically significant.
In addition, population comparisons were performed using data gathered from the National Health and Nutrition Examination Survey (NHANES) database from 2003-2016 (N = 4,370). Using the NHANES data, similar cohorts of college-aged respondents across ethnicity, gender, and income were created. Sample weighting and analysis of the NHANES data were performed using Python (version 3.6.5). These subsets of NHANES data were weighted using the provided sample weights from each year and adjusted to fit the age ranges of the cohorts and to create nationally-representative comparisons to the Furman University data. For each of the cohorts, similar analyses were performed to provide cholesterol values and 95% confidence intervals (TC, HDL and triglycerides). The NHANES data was then compared to the Furman University results using the appropriate confidence intervals and the degree to which they overlap ( Figure 1). P-values <0.05 were 7 considered statistically significant.

Furman University Students
A total of 4423 college students who attended Furman University between 2003-15 were included in the analysis (Table 1). African-American students had significantly lower VLDL and triglyceride levels, with significantly higher HDL levels compared to other race/ethnicities. African-Americans also demonstrated greater percentages of athletic involvement overall (independent of healthier cholesterol values) with 18% involvement versus 7.3% total student involvement (  (Table 3). African-Americans demonstrated significantly lower triglycerides. Overall, Furman students demonstrated healthier lipid averages than the NHANES age-matched comparison group in all categories, regardless of socioeconomic background or race/ethnicity (see Figure 1). 8 Our study suggests that most college-age students, as expected, will demonstrate normal lipid values throughout adolescence; and, based on the likely rate of rise over time, may maintain similar patterns through youthhood based on the low prevalence of abnormal values. A small fraction of our database showed evidence of hypercholesterolemia per NHLBI guidelines 12  With regard to the higher HDL-C values in Furman college African-American students, a similar Multiple Risk Factor Intervention Trial has also shown African-American males demonstrating on average 10 mg/dL higher HDL-C levels and an inverse relationship between HDL-C and SES status in black men vs white men (while SES status and HDL-C was directly correlated in white men) 13 . While more research is needed in this area to confirm these findings, this may give support to the concept that African-American students who are physically active and subsequently more health-conscious are also more 9 likely to avoid CVD than other less active ethnic or minority individuals. Regardless, additional research on which socioeconomic factors contribute to these perplexing trends in cholesterol health disparities worldwide is clearly indicated due to these findings in the current population. This should also raise awareness of the possibility of different lipid thresholds existing for various ethnicities; in other words, we cannot disprove a continued higher risk for clinical CVD at an early age for certain minority populations despite normal youth lipid screening test results.

Discussion
Previous studies in cholesterol research demonstrate that low socioeconomic status (SES) and minority ethnicity are established risk factors for poor cholesterol health, which is impacted by both quantifiable and nonquantifiable determinants 14,15,16,17,18,19,20 .
Quantifiable risk factors are cholesterol values (including non-fasting triglyceride levels), hemoglobin A1c levels, smoke exposure 21 , family history of stroke, diabetes, or obesity (specifically for African-American youths) 22 , and [from highest to lowest odds ratios] single-living, non-white status, low income, and low education 23 . Nonquantifiable risk factors are much broader and include language barriers and lack of healthcare access 4 , low and medium educational levels 16 , having a negative affect in familial interactions 24 , and poor health promotion behaviors including low knowledge of personal risk indicators for CVD 25,26 . Minority status also is recognized as an independent risk factor for high blood pressure, heart attack and stroke 27 , lower rates of moderate exercise 28  One limitation of our study may exist as more Caucasian students attend Furman than other average university populations; however, our large sample size allowed us to include students from diverse socioeconomic and minority backgrounds with strong statistical associations within each group and lipid subcategory. We also recognize some discrepancies exist inherent within comparisons between the NHANES data and our Furman University data, including the assumption that our high financial need category in Furman's database is comparable to the NHANES low income category (and vice versa).
However, we have included the various salaries for each group in our data tables below to avoid confusion in SES classification. A strength of our study is the detailed comparison of various types of students within the Furman database such as athlete and non-athlete status and the inclusion of several levels of socioeconomic status, which is often absent from similar lipid studies. Another strength is the comparison of our original data with an age-matched control database from NHANES, which allows for unique verification of cholesterol trends between distinct groups of youths.
As described above, adult-onset cardiovascular disease continues to be identified (and treated) by clinicians worldwide using routine lipid screening protocols in adolescence and beyond. However, even with current technological advancements in cardiovascular imaging and updated lipid screening recommendations, less is known about how to identify a younger and more insidious version of lipid disease that exists with little to no clinical symptoms in adolescence. To complicate future research in lipid screening even further, other lipid study efforts demonstrate that several nonnumerical factors including psychosocial, social, and family of origin characteristics are also related to lipid health, and that healthy behaviors in these areas maintain some degree of lifelong protection from CVD 33 . Frankly, the literature seems to lack enough well-described data or demographic associations to fully circumscribe the bulk of youth lipid disease as compared with CVD in adult populations. As all clinicians surely desire to identify all children and youths with serious lipid disease, we do not know which adults have lived with long-standing atherosclerosis or lipid disease since childhood unless routine lipid screening in all children is strongly recommended by national guidelines. Otherwise delayed diagnosis and treatment is overdue when initiated at age 35 or even beyond (per USPSTF guidelines) 34 , and CVD morbidity and mortality will continue to impact these individuals for many years unless change is implemented.
Other researchers have shown that black ethnicity is associated with healthier lipid values, most notably the Bogalusa study demonstrating higher HDL and lower triglycerides in blacks vs whites 35 . A more recent study by Sumner et al also demonstrated similar findings and also mentions that blacks have higher rates of cardiovascular disease and diabetes but paradoxically are diagnosed less often with metabolic syndrome. This is likely due to the requirement that low HDL and high triglycerides make up two-fifths of the criteria for diagnosing metabolic syndrome (along with central obesity, hypertension, and fasting hyperglycemia), which blacks are less likely to have despite worse disease outcomes 36 . Thus the data our study unearthed likely represents a similar relationship and does not necessarily represent cardioprotective effects from seemingly healthier lipid values, although more research in the age group from our study is needed to verify these trends as compared with adults.

Conclusions
College-age socioeconomic status has little impact on lipid parameters within our study of private liberal arts students, but college enrollment may represent a marker for healtheir lipid values when compared with age-matched individuals (from the NHANES database).
Furthermore, while many pediatric and cardiovascular-specific lipid screening algorithms (AAP, NHLBI) encouarge primary care providers to routinely assess lipid health in patients of all backgrounds and risk factors, many young individuals at risk for future cardiovascular disease may be missed if non-pediatric screening tools (USPSTF, etc.) are used as these models do not recommend routine adolsecnt lipid screening.
Finally, we hope our findings encourage other universities to adopt similar practices as Furman's Health Sciences Department in lipid screening, in conjunction with the practical application of other wellness techniques to encourage youth health. Our data suggest that private liberal arts college students may have less CVD risk than an age-matched comparisons, yet some students may demonstrate significant morbidity if they remain untreated until visiting a primary care provider as a middle-aged adult. We hope to further expand this research project to help Furman University and other schools implement a similar model to predict future cardiovascular health issues in college students with immediate research-based data according to their SES/ethnicity-associated predispositions.

Declarations
Ethics approval and consent to participate: Not applicable Consent for publication: Not applicable Availability of data and material: The data that support the findings of this study are available from Furman University but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available.
Data are however available from the authors upon reasonable request and with permission of Furman University (along with consent from collaborating partners with USC-School of Medicine-Greenville).
Competing interests: The authors declare that they have no competing interests.
Clinical Trial Registration (if any): none Funding: No external funding for this manuscript. All authors have indicated they have no financial relationships relevant to this article to disclose. All authors have indicated they have no potential conflicts of interest to disclose.

Authors' contributions:
Dr. Hudson and Dr. Trilk conceptualized and designed the study, drafted the initial manuscript, produced the figure(s) and reviewed and revised the manuscript.
Dr. Feigenbaum (with Furman University) designed the data collection system, collected data, and reviewed and revised the manuscript.
Mr. Patil performed all statistical analyses for the Furman lipid data and also generated the associated Furman lipid data table and reviewed the manuscript for important statistic and intellectual content.
Mr. Ewing and Ding performed all statistical analyses for the NHANES lipid data and produced the associated NHANES lipid data table and also reviewed the manuscript for important statistic and intellectual content.
All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.