July 8, 2015
Albert L. Siu, M.D., M.S.P.H.
Chair, United States Preventive Services Task Force
540 Gaither Road
Rockville, MD 20850
Re: Draft Research Plan for Cardiovascular Disease: Risk Assessment Using Nontraditional Risk Factors
Dear Dr. Siu:
The Academy of Nutrition and Dietetics (the "Academy") appreciates the opportunity to submit comments to the United States Preventive Services Task Force (USPSTF) related to its June 11, 2014 Draft Research Plan for Cardiovascular Disease: Risk Assessment Using Nontraditional Risk Factors ("Draft Research Plan").1 With over 75,000 members comprised of registered dietitian nutritionists (RDNs),2 nutrition and dietetic technicians, registered (DTRs), and advanced-degree nutritionists, the Academy is the largest association of food and nutrition professionals in the United States committed to improving the nation’s health through food and nutrition across the lifecycle. The Academy's members independently provide nutrition care services, including Medical Nutrition Therapy (MNT),3 to individuals with a wide variety of disease states and chronic conditions, including cardiovascular disease (CVD) and risk factors of CVD.
The Draft Research Plan could have significant impact for individuals at risk of CVD if the systematic evidence review is structured to gather existing evidence of the benefits of addressing nontraditional risk factors through intensive behavioral counseling and MNT. In addition, the Draft Research Plan can be modified to measure additional indicators due to the multiple factors involved in the development of heart disease. Thus, to optimize the chances that the USPSTF detects and recognizes substantial benefits obtained through nutrition care services, the Academy suggests the below additional considerations for the Draft Research Plan.
I. Choice of Biomarkers and Other Factors for Risk Assessment
The Academy recommends that USPSTF explicitly adopt the biomarker evaluation process recommended by the Institute of Medicine in the 2010 report, Evaluation of Biomarkers and Surrogate Endpoints in Chronic Disease.4 This three-step process outlined in the excerpt below ensures that, for each biomarker, the tests used are dependable, the prognostic value and role in the disease process are understood, and the applications in practice produce the desired results.
- Analytical validation: analyses of available evidence on the analytical performance of an assay;
- Qualification: assessment of available evidence on associations between the biomarker and disease states, including data showing effects of interventions on both the biomarker and clinical outcomes; and
- Utilization: contextual analysis based on the specific use proposed and the applicability of available evidence to this use. This includes a determination of whether the validation and qualification conducted provide sufficient support for the use proposed.
While it is clear from the current research plan and the results of USPSTF's 2009 review on this topic that these concepts are already to be considered, the adoption of the IOM's evaluation process and devotion of key questions to each step would serve to promote the sound use of biomarkers in healthcare recommendations and to ensure that each aspect is thoroughly evaluated. Additionally, the explicit use of the three step process would help to elucidate the vital difference between risk factor and surrogate endpoint status for the biomarkers to be evaluated and support the appropriate application of the recommendations resulting from this research.
The Academy believes that traditional CVD risk assessment measures should also include LDL-cholesterol. Although LDL-cholesterol is not part of the Framingham or Pooled Cohort Equation, it is widely recognized as a better marker of atherosclerotic risk than total- or HDL-cholesterol, and is certainly a part of "traditional" approach to assessing and monitoring risk.5 The USPSTF should also include nontraditional CVD risk assessment measures that are considered good quality biomarkers of risk of CVD (or that are related to its development) and that can be reasonably targeted by lifestyle changes in this review's target population of adults without known CVD.
Specific processes and conditions involved in CVD the Academy believes should be reflected in assessment include:
Several measures related to atherosclerosis are presently included in the Draft Research Plan; coronary artery calcium and ABI are highly appropriate measures and reflect the 2013 AHA/ACC Guideline on the Assessment of Cardiovascular Risk.6 In addition, we recommend several additional markers for possible inclusion in this Draft Research Plan:
- Non-HDL cholesterol. Although current research supports LDL's effectiveness, it generally suggests that non-HDL cholesterol is likely a better predictor of CVD risk.
- ApoB and LDL-P. Recent systematic reviews commend their validity as markers of risk. Additional research into their validity could enhance use in general clinical practice.
Measures related to chronic inflammation should be included in the Draft Research Plan. Hs-CRP is a primary biomarker of inflammation and its inclusion in the Draft Research Plan appropriately reflects relevant guidelines for assessment of risk.8 Hs-CRP is also a marker amenable to change through diet, which enables the evidence review to detect and recognize the substantial benefits from diet and nutrition interventions conducted by highly qualified practitioners, such as RDNs. In addition, since inflammation is now recognized as a sentinel process in cardiovascular disease development,9 the Academy suggests consideration of at least one additional inflammatory biomarker that offers the best combination of predicting CVD risk that is amenable to change through diet and lifestyle choices.
C. Insulin Resistance
Evidence is reasonably strong that even without diagnosis of diabetes or prediabetes, elevated fasting and post-prandial insulin levels appear to be associated with development of cardiovascular disease.10> In these early stages of dysfunction, insulin resistance can be reduced successfully by diet, physical activity and weight loss interventions, making it a relevant measured outcome.11 We note that consideration of HOMA-Insulin or c-peptide as biomarkers of insulin resistance could be included, but that fasting insulin levels would not be an appropriate substitute for these additional biomarkers.
D. Metabolic Syndrome
Metabolic syndrome doubles the risk of cardiovascular mortality and risk of stroke.12 It is amenable to change through diet interventions, physical activity, and modest weight loss (especially loss of visceral fat in the abdomen).13 The traditional risk factors included in the Draft Research Plan include blood pressure, diabetes, and HDL, which are some of the individual components of metabolic syndrome. However, not all relevant metabolic syndrome components are included. Further, current research suggests that the presence of metabolic syndrome may convey additional meaning beyond the individual components used to define it.14
Therefore, the Academy recommends including diagnosis of metabolic syndrome as an additional nontraditional risk factor. Appropriate use likely requires agreement of standards to metabolic syndrome and whether it differs from the existing interim international definition in place.15
1. Dysfuctional Adipose Tissue
Dysfunctional adipose tissue appears to be at the heart of metabolic syndrome. This requires moving beyond simply looking at measures of obesity like body mass index (BMI), since studies show that people of the same BMI can have vastly different body composition. We note that the Draft Research Plan does not include any measure of weight or body fat. Thus, while BMI could be used as a measure, if it is so used, it should be analyzed as BMI itself, not used after categorization as either normal, overweight, or obese. The 5 to 10% weight loss that is a realistic goal of nutrition interventions and has been shown effective in reducing various markers of metabolic and cardiovascular health may only reflect a minimal change in BMI.16 It may be preferable to include a measure of weight change either in absolute or percentage terms and perhaps continue the follow-through of further documenting change in risk factors and accomplishment through lifestyle change.
Although not necessarily an accurate measure of metabolically active visceral adipose tissue, many researchers consider a measure of weight (such as BMI) combined with waist circumference a better measure of cardiometabolic risk than relying on weight or BMI alone.17 The Academy suggests that if studies include waist circumference or change in waist circumference, that it be included in analysis of the data.
2. Serum Triglycerides
Elevated serum triglycerides are one component of metabolic syndrome and are associated with post-prandial lipemia and exposure of arteries to atherogenic lipoproteins,18 making them an important CVD risk factor. Triglyceride levels are also responsive to the lifestyle changes and interventions the USPSTF intends to review. Serum triglyceride change would be reflected in change of non-HDL cholesterol (if the Academy's recommendation to include that measure is accepted), but we note that many studies (especially those conducted less recently) may not include the non-HDL cholesterol. Thus, including triglycerides could capture this potential indicator of improved cardiovascular health.
E. Food Insecurity
The Academy recommends that the USPSTF include a score from a validated food security scale in the non-traditional risk factor assessment. Research has shown that food insecurity is associated with a variety of diet-sensitive chronic diseases, including cardiovascular disease. These associations are largely mediated by a poor quality diet – one rich in inexpensive, energy-dense, and nutrient-poor processed foods – as well as lifestyle factors that contribute to chronic disease. One study showed that food insecure adults aged 30 to 59 have an increased 10-year predicted risk for CVD.19 Researchers examined 10,455 adults aged 20 and older from the National Health and Nutrition Examination Surveys (NHANES) from 2003-2008 and found that participants with very low food security had a 10-year predicated CVD risk that was over 20% greater than food secure participants.
The increased CVD risk profile associated with food insecurity is attributable to higher levels of a cluster of CVD risk factors. Research demonstrates that food insecure individuals have higher rates of dyslipidemia, type 2 diabetes, poor glycemic control, metabolic syndrome, and obesity, with particularly strong associations among women. Women who are marginally food insecure were more likely to have high LDL cholesterol and triglyceride levels as well as abnormal triglyceride/HDL ratios.20 Food insecurity was associated with low HDL cholesterol levels in women.21 Food insecurity has also been found to correlate with hypertension, hyperlipidemia, diabetes and inadequate glycemic control, independent of obesity status, leading to the conclusion that, "food insecurity is associated with cardiovascular risk factors."22 Food insecure households have also been found to be at greater risk for metabolic syndrome, which includes an increased waist circumference, high blood pressure (prehypertensive/ hypertensive), low HDL cholesterol, high triglycerides, and impaired glycemic control (prediabetes or diabetes).23 Food insecure individuals also have been demonstrated to have high levels of inflammation, as demonstrated by elevated C-reactive protein profiles, which correlates with chronic disease.24
II. Nature and Classification of Interventions
A. Intensity and Duration
Since 1996, the USPSTF and affiliated researchers have undertaken a series of evidence reviews and systematic evidence updates analyzing available studies related to dietary interventions and counseling to promote a healthy diet. These reviews identify intensity of intervention (minutes in a session) as the most significant factor; after intensity, the most consistent factors of effective interventions are the specialized qualifications of the provider and the provision of services at a venue being outside the primary care setting. The largest changes in diet come from interventions conducted by a registered dietitian nutritionist or other specially trained practitioner conducted outside of the primary care setting. Thus, search terms used in the literature search are important to make sure that applicable dietary interventions are included. Some may be referred to in the literature using other terms. Therefore we suggest that search terms include "medical nutrition therapy" and "dietary counseling," including any counseling provided by "dietitians," "dieticians," "nutritionists," or that includes a dietitian on the health care intervention team. We encourage the USPSTF to incorporate studies and recommendations from recently published evidence reviews and guidelines and to differentiate protocol-based behavioral counseling from more complex individualized Medical Nutrition Therapy, the latter of which is a recognized necessary and effective component of care for individuals with multiple chronic conditions (including prevention of CVD).
Description of interventions is important for USPSTF's evaluation of results, as noted in the USPSTF recommendation regarding behavioral counseling for cardiovascular disease prevention among adults with known risk factors and numerous other reports and analyses of research. In short, intensity of interventions often has strong influence on outcomes. The Academy supports the identification of comparison groups in this USPSTF research plan as no intervention, minimal intervention, and attention control.
It is important that the USPSTF not group together all interventions in conclusions. That is, the Draft Research Plan should ensure that outcomes from a one-time half-hour group class could not be considered equally alongside a 12-week program of weekly hour-long behavioral sessions that also includes supervised weekly exercise. Past reports have made note of how differences in intensity influence results, for example noting that low intensity interventions are ineffective for an outcome for which a medium or high intensity intervention is effective. If, as would be expected given other analyses and reviews, more intense interventions are more effective relevant to the research questions posed here, this has important implications for quality of patient care that is supported, and for the opportunity for RDNs to do the more intense interventions we are trained to provide.
We note that for these purposes, intensity of interventions has relatively standard definitions related to number of sessions and frequency. Variations could include duration of sessions and the time period over which an intervention extends. In addition, when data is reviewed, the Academy recommends that separate note be made of diet, physical activity or the combination, and also whether the intervention did or did not target weight loss. Many of these biomarkers are sensitive to excess body fat and change in weight. Some studies have found that without targeting weight in those who are overweight and obese, results are more limited. On the other hand, if a program can target healthy habits (eating and activity) without specifically targeting weight loss, that would set precedent for developing (and reimbursement for) interventions to reduce CVD risk.
B. Provider Type
The USPSTF has repeatedly recommended intensive behavioral [dietary] counseling for various disease states and conditions with what appeared to be clear recommendations that primary care providers are able to refer patients to RDNs and other qualified experts outside of the primary care setting to provide and deliver these services. Frustratingly, despite the USPSTF's intention for its recommendations to be primary care relevant and to include referral, CMS indicates to us that USPSTF recommendations are only "for primary care clinicians," in part because that is who comprises Task Force members. As a result, the Task Force's recommendations for some of the nation's most pressing problems cannot be implemented in Medicare or public and private insurance that follows CMS's lead. The Academy respectfully requests a meeting at your earliest convenience to discuss these hurdles in implementing Task Force recommendations.
As the USPSTF has found, primary care providers are often limited in their time, training, and skills to conduct the medium or high-intensity interventions that are scientifically proved to be the most effective in producing the largest, most lasting results. It is both cost-effective and efficient to have primary care provider-driven referrals of patients with risk factors for CVD to practitioners skilled in conducting dietary interventions who practice both inside and outside of primary care settings.
The Academy and God's Love We Deliver, who also supports this comment, recognizes the complexity of this topic and offer our assistance and evidence analysis as USPSTF begins its systematic review of the evidence. Please contact either Jeanne Blankenship at 202/775-8277, ext. 1730 or by email at firstname.lastname@example.org or Pepin Tuma at 202/775-8277, ext. 6001 or by email at email@example.com with any questions or requests for additional information.
1 U.S. Preventive Services Task Force website. Draft Recommendation Statement: Behavioral Counseling to Promote a Healthy Diet and Physical Activity for Cardiovascular Disease Prevention in Adults with Known Risk Factors. Accessed May 29, 2014.
2 The Academy recently approved the optional use of the credential "registered dietitian nutritionist (RDN)" by "registered dietitians (RDs)" to more accurately convey who they are and what they do as the nation’s food and nutrition experts. The RD and RDN credentials have identical meanings and legal trademark definitions.
3 "Medical Nutrition Therapy is an evidence-based application of the Nutrition Care Process. The provision of MNT (to a patient/client) may include one or more of the following: nutrition assessment/re-assessment, nutrition diagnosis, nutrition intervention, and nutrition monitoring and evaluation that typically results in the prvention, delay, or management of diseases and/or conditions." CITE definition of terms list
4 Evaluation of Biomarkers and Surrogate Endpoints in Chronic Disease. IOM (Institute of Medicine). 2010. Washington, DC: The National Academies Press.
5 Expert Dyslipidemia Panel of the International Atherosclerosis Society Panel members. An International Atherosclerosis Society Position Paper: Global Recommendations for the Management of Dyslipidemia—Full Report. J Clin Lipidol. 2014 Jan-Feb;8(1):29-60.
6 Goff DC, Lloyd-jones DM, Bennett G, et al. 2013 ACC/AHA Guideline on the Assessment of Cardiovascular Risk: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation. 2014;129(25 Suppl 2):S49-73.
7 Jacobson TA, et al. National Lipid Association Recommendations for Patient-centered Management of Dyslipidemia: Part 1 - Executive Summary. J Clin Lipidol. 2014 Sep-Oct;8(5):473-88.
8 Goff DC, Lloyd-jones DM, Bennett G, et al. 2013 ACC/AHA Guideline on the Assessment of Cardiovascular Risk: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation. 2014;129(25 Suppl 2):S49-73.
9 Golia, E., et al., Inflamation and Cardiovascular Disease: From Pathogensesis to Therapeutic Target. Curr Atheroscler Rep. 2014 Sep;16(9):435. doi: 10.1007/s11883-014-0435-z.
10 Reaven GM. Insulin Resistance, the Insulin Resistance Syndrome, and Cardiovascular Disease. Panminerva Med 2005;47:201-210.
11 Mason, Caitlin et al. "Dietary Weight-Loss and Exercise Effects on Insulin Resistance in Postmenopausal Women." American Journal of Preventive Medicine 41.4 (2011): 366–375. PMC. Web. 8 July 2015.
12 Alberti KG, Eckel RH, Grundy SM, et al. Harmonizing the Metabolic Syndrome: A Joint Interim Statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation. Oct 20 2009;120(16):1640-1645.
13 Jas Mottillo S, Filion KB, Genest J, et al. The Metabolic Syndrome and Cardiovascular Risk a Systematic Review and Meta-analysis. J Am Coll Cardiol. Sep. 28 2010;56(14):1113-1132.
14 Ratto, E., et. al., Metabolic Syndrome is Associated with Left Ventricular Dilation in Primary Hypertension; Journal of Human Hypertension, (25 June 2015) doi:10.1038/jhh.2015.67.
15 Alberti KG, Eckel RH, Grundy SM, et al. Harmonizing the Metabolic Syndrome: A Joint Interim Statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation. 2009;120(16):1640-5.
16 Goff DC, Lloyd-jones DM, Bennett G, et al. 2013 ACC/AHA guideline on the Assessment of Cardiovascular Risk: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation. 2014;129(25 Suppl 2):S49-73.
17 Alberti KG, Eckel RH, Grundy SM, et al. Harmonizing the Metabolic Syndrome: A Joint Interim Statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation. Oct 20 2009;120(16):1640-1645.
18 Miller, M., et al., Triglycerides and Cardiovascular Disease: A Scientific Statement From the American Heart Association. Circulation. 2011;123:2292-2333; originally published online April 18, 2011; doi: 10.1161/CIR.0b013e3182160726.
19 Ford ES (2013). Food Security and Cardiovascular Disease Risk Among Adults in the United States: Findings from the National Health and Nutrition Examination Survey, 2003-2008. Preventing Chronic Disease 10: 130244. doi: http://dx.doi.org.10.5888/pcd10.130244.
20 Tayie FA, Zizza CA (2009). Food Insecurity and Dyslipidemia Among Adults in the United States. Preventive Medicine 48(5): 480–485.
21 Shin, J.-I., et al., (2015). Food Insecurity and Dyslipidemia in a Representative Population-based Sample in the US. Preventive Medicine, http://dx.doi.org/10.1016/j.ypmed.2015.05.009.
22 Seligman HK, Laraia B, Kushel MB (2010). Food Insecurity is Associated with Chronic Disease Among Low-income NHANES Participants. Journal of Nutrition 140(2): 304–310.
23 Parker ED, Widome R, Nettleton JA, Pereira MA (2010). Food Security and Metabolic Syndrome in U.S. Adults and Adolescents: Findings From the National Health and Nutrition Examination Survey, 1999–2006. Annals of Epidemiology 20 (5):364–370.
24 Gowda C, Hadley C, Aiello AE (2012). The Association between Food Insecurity and Inflammation in the US Adult Population. American Journal of Public Health 102(8): 1579–1586.