Nutrition Research: Restoring Respect to a Field Under Siege

An Interview With Frank Hu

Interviewer: Tricia Ward; Interviewee: Frank B. Hu, MD, PhD


January 23, 2019

Nutrition research has been under siege. In addition to the perennial low-fat, low-carb debates, there were calls for a complete overhaul of the field[1] by John Ioannidis, MD, DSc. At the 2018 American Heart Association meeting, Frank B. Hu, MD, PhD, gave the Ancel Keys Memorial Lecture. We asked the nutritional epidemiologist about healthy eating and the way forward for his field.

Frank B. Hu, MD, PhD | Medscape: Before we wade into the areas of controversy, what are the key areas of consensus on dietary advice?

Hu: It is generally recognized that there is no one-size-fits-all diet that is best for everyone. One can combine foods in a variety of flexible ways to achieve healthy dietary patterns, which can be adapted to meet individual health needs, food preferences, and cultural traditions. For most people, a healthy diet typically includes:

  • Higher amounts of fruits, vegetables, whole grains, legumes, and nuts;

  • Lower amounts of refined grains and red and processed meats;

  • Low consumption of added sugar, especially sugar-sweetened beverages;

  • Reduced intake of sodium; and

  • Appropriate portion size control and balancing caloric intake with physical activity to manage weight.

There is growing acceptance that our dietary patterns need to consider both human health and environmental impact. There is also increasing consensus that unhealthy food environments can play a large role in shaping individuals' food choices. Therefore, population-based strategies, such as public policies, are needed to improve the food environment by making healthy and sustainable choices the easy, affordable, and default options. | Medscape: What needs to change to improve the standing of nutritional epidemiology? Should academics/professional societies develop a set of standards for nutritional research?

Hu: It is difficult to come up with a checklist of "standards" for good nutritional epidemiologic research, because different research questions require different study designs, dietary assessment methods, and statistical approaches. Like any other field, we should strive to improve scientific rigor and reproducibility of nutritional epidemiologic studies. Solid evidence is built on high-quality studies in humans that typically include the following features:

  • Large numbers of participants and sufficient statistical power;

  • High follow-up rates and good dietary compliance (in intervention studies);

  • Carefully designed and validated dietary assessment tools (repeated measures are desirable);

  • Carefully conducted statistical analyses that take into account potential confounding factors;

  • Confirmation by other studies (reproducible results);

  • Supportive evidence from short-term controlled feeding studies with biochemical or physiologic outcomes; and

  • Careful interpretation of data in the context of available evidence and limitations of the study design. | Medscape: What about the role of journal editors in publishing weak research and researchers/media in exaggerating such data?

Hu: Authors, journals, and the media should avoid sensationalism and exaggeration of research findings. We should also keep in mind that newer studies are not necessarily better or more reliable than previous ones, and it is always important to look at the totality of the evidence. | Medscape: In your lecture, you equated food assessment in nutritional epidemiology to blood pressure measurement. Can you elaborate for our readers?

Hu: Analogous to blood pressure measurement, multiple measurements of diet (using either self-reported methods or biomarkers) can reduce noise in the data and improve accuracy of dietary assessment.

Among the self-reported methods, a single 24-hour recall is the least accurate[2] owing to day-to-day variations in dietary intakes, making it largely unsuitable for examining associations between diet and disease. The 24-hour recall, however, is still useful in estimating population averages or trends of certain nutrients and foods over time.

Multiple-week diet records provide the most detailed information on diet, but this method is expensive and very burdensome for participants, and therefore is not feasible in most large population-based studies. Carefully designed and validated food-frequency questionnaires (FFQs) can provide reasonably accurate long-term data on most dietary factors, and repeated FFQs over time can be used to reduce measurement errors and improve accuracy. Objective biomarkers measured in blood or urine samples are useful in assessing intake of some nutrients and foods. However, sensitive and specific biomarkers are not available for most nutrients and foods.

The main takeaway is that there is no perfect method to measure diet in free-living populations. Nutritional research needs to consider a combination of multiple methods for data collection, including repeated self-reported measures and biomarkers as well as smartphones and sensors, depending on the study design and characteristics of the study population. | Medscape: In the JAMA viewpoint on nutritional epidemiologic research,[3] the reduction in trans fats is cited as a successful example of nutritional epidemiology affecting policy, but weren't trans fats created in response to nutritional epidemiology studies demonizing saturated fats?

Hu: The technology used to produce partially hydrogenated vegetable oils that are high in trans fat was developed in the [early] 1900s. The use of these processed vegetable fats became more widespread in the 1960s to replace animal fats, first because of lower costs and then because of purported health benefits. It turns out that trans fat is worse than saturated fat in terms of heart disease risk.

On the basis of evidence from intervention trials and epidemiologic studies on trans fat and heart disease risk, the US Food and Drug Administration required trans fat labeling in 2006 and banned trans fat in 2018. These regulations have led to an approximate 70% reduction in trans fat intake in the US population.[4] This is a huge improvement in diet quality. | Medscape: You support restrictions on saturated fat (up to 10%, per the 2015 Dietary Guidelines for Americans) despite data suggesting that saturated fat intake may not raise risk for cardiovascular disease (CVD).[5]

Hu: The fact that trans fat is worse than saturated fat does not mean that saturated fat is beneficial or benign in terms of CVD risk. Likewise, the fact that refined starch/added sugars are harmful does not exonerate saturated fat.

The key question we should ask when assessing the effects of a nutrient or food is, "Compared with what?" Compared with trans fat, saturated fat is better, but compared with unsaturated fats from olive oil and other vegetable oils,[6,7] nuts/seeds, avocados, and seafood, saturated fat increases risk for CVD.

Many epidemiologic studies have found no significant association between saturated fat intake and coronary heart disease risk. In these studies, saturated fat was typically compared with carbohydrates,[8] which are mostly refined carbs and added sugars. In terms of dietary practice, it's not a good idea to cut back on saturated fat but eat more white bread, bagels, sugary foods, and beverages. Instead, saturated fat should be replaced with unsaturated fats and healthy carbs from legumes and whole grains.[9]

It seems that the debate on low-fat versus low-carb diets will not end anytime soon, but such a debate is not meaningful unless the food sources of fats and carbohydrates are defined. Not all low-fat or low-carb diets are created equal. No matter what dietary pattern you follow, you should pay attention to the food sources of fats and carbs. | Medscape: Can you address two potentially controversial inclusions in your definition of a healthy dietary pattern, firstly coffee? Do you think the data are strong enough to say that coffee is actually healthy (as opposed to not harmful)?

Hu: There is compelling evidence that moderate consumption of coffee (three to five cups per day) is not associated with increased risk for chronic diseases, such as heart disease and cancer.[10] The same line of evidence also indicates that coffee consumption is associated with a lower risk for type 2 diabetes, heart disease, and some cancers.

Current dietary guidelines recommend that moderate coffee consumption can be included as part of a healthy diet and lifestyle. The main message is that if you enjoy coffee, you can continue to do so. But if you prefer tea, there is no compelling reason to switch from tea to coffee. | Medscape: Presumably the three to five cups per day are not venti lattes.

Hu: No. The standard portion size for a cup of coffee is 8 oz. Three to five 8-oz cups would provide about 400 mg/day of caffeine. Many coffee shops serve portion sizes that are two to three times larger than this standard, and have specialty coffee drinks that are loaded with calories, sugar, and saturated fat. In addition to paying attention to portion size, it is also important to limit unhealthy additions to coffee, such as sugar and cream. | Medscape: You also include moderate alcohol in a healthy diet. The World Health Organization's International Agency for Research on Cancer says that no amount of alcohol is safe and that alcohol is a causative agent in some cancers.

Hu: It has been long recognized that alcohol consumption is associated with increased risk for cancer in a dose-response manner. A key question is how to balance health risks and potential benefits of moderate alcohol consumption when making recommendations. Current dietary guidelines strongly discourage excessive alcohol consumption and do not recommend that individuals who do not drink alcohol start drinking for any reason. | Medscape: Some argue that the supporting data for alcohol and CVD are largely from research funded by the alcohol industry[11] or that moderate alcohol consumption is merely a marker of a healthy lifestyle.[12]

Hu: The weight of evidence from hundreds of studies continues to suggest that moderate alcohol consumption is associated with lower risk for heart disease, despite some recent controversies.

Most studies took into account other diet and lifestyle factors and were not funded by the alcohol industry. The consistency of evidence is remarkable.

It is also important to keep in mind that alcoholic beverages contain substantial calories, which could play a role in weight gain. | Medscape: Your talk addressed the obesogenic environment we live in. Are you optimistic or pessimistic about reversing it?

Hu: I'm still optimistic that the obesogenic environment can be reversed, but it is not going to be easy. The good news is that there is an increasing recognition of the importance of societal and public policy factors in shaping our eating habits, not just among academic researchers but also among policy-makers and industry leaders.

However, political leadership is critical to developing and implementing policy solutions that have the potential to transform our food environment and change our cultural norms about healthy eating. It took more than five decades for antismoking efforts to achieve the success in reducing cigarette smoking that we see today in the United States. Given its complexity, it may take even longer (I hope not!) and require even greater efforts to halt and reverse the obesity epidemic. | Medscape: Finally, what advice do you have for those who are confused by nutrition research?

Hu: When you see a sensational or attention-grabbing headline about a nutrition study, take a deep breath and ask how reliable the study is and how it fits with the rest of the literature. Scientific research is an evolving process, and contradictions between published studies are inevitable and a healthy part of this process. However, these contradictions can cause confusion among the public, especially considering that media reports and Internet bloggers tend to oversimplify nutrition studies. So it is important to be skeptical when reading sensational headlines touting quick-fix diets or wildly exaggerated and provocative opinions on nutritional studies.

Follow Tricia Ward on Twitter: @_triciaward

For more Cardiology, follow us on Twitter: @theheartorg


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.