Until recently, and still today in low-income countries, undernutrition during pregnancy and the first years of life was a major cause of mortality. However, in recent decades, both in the United States and globally, noncommunicable diseases (NCDs) account for the majority of premature deaths. In analyses of the overall burden of disease, dietary factors were identified as the most important causes in the United States and worldwide.1 Despite this, physicians and other healthcare providers are taught little about nutrition in medical school or fellowship training.2 Not surprisingly, in conventional medical practice almost no attention is given to knowing what a patient is eating or providing dietary guidance that has the potential to improve dramatically their long-term health. To counter widespread misinformation and confusion about diet and health, this chapter describes what we know about the elements of a healthy diet and how these elements can be combined into an overall dietary pattern for the prevention of major illness and promotion of well-being. A brief section considers ways that this knowledge can be integrated into preventive healthcare.
Origins of Modern Concepts on Diet and Health
Although diet has been recognized since the time of Hippocrates as an underlying determinant of health, more recent evidence that diet might be important in the prevention of cardiovascular disease and cancer came from studies comparing rates of disease in different countries. Some of the most important evidence emerged from the 7-Countries Study conducted by Dr. Ancel Keys and colleagues during the 1950s and 1960s.3 These investigators, for the first time, documented incidence and mortality rates of coronary heart disease (CHD) using standard diagnostic criteria in 14 populations of about 1,000 men in each of 7 different countries. They found an approximately 10-fold difference in rates, with the highest being in Finland and the lowest in Crete. One possible explanation for these huge difference could be genetic factors, but other scientists investigating heart disease among migrants from low-incidence countries to the United States found that the migrants rapidly adopted the rates of heart disease among European Americans,4 firmly rejecting genetic factors as the explanation for the high rates of heart disease in the United States and northern Europe. This simple but critical finding meant that CHD, the number one cause of death in the United States, was potentially preventable if the causal factors could be identified and modified.
In their quest to identify possible explanations for the large variation in CHD rates, Keys and colleagues noted that these rates correlated strongly with intake of saturated fat but not total fat; those with the highest total fat intake (about 40% of calories) included both Finland, where dairy fat was the dominant source, and Crete, where olive oil was the dominant source.3 Although these correlations with diet raised important questions, they could have been explained by confounding factors, such as differences in smoking, physical activity, or other aspects of diet, thus indicating the need for more detailed studies. Controlled feeding studies examining changes in blood lipids over a few weeks provided support for saturated fat contributing to heart disease because replacement of carbohydrate intake with saturated fat increased blood total cholesterol. However, these studies did not support total fat being responsible because polyunsaturated fat reduced total blood cholesterol and monounsaturated fat appeared neutral.5
During this period, other epidemiologists were investigating rates of breast, colon, prostate, and other cancers around the world. They also found great variation in rates among countries, as much as 8- to 10-fold.6 Also, like heart disease, the rates in populations migrating from low- to high-risk countries converged with, and sometimes overshot, the rates in European Americans, although for breast cancer the catch-up occurred over several generations.7 Again, strong correlations were seen between intakes of total or animal fat and rates of these cancers.
Although these crude correlations should have been interpreted cautiously because many potentially confounding factors existed, they were widely interpreted as causal.8 As a result, beginning in the early 1980s, dietary recommendations in the United States and worldwide emphasized reduction in total fat intake. The cardiovascular prevention world went along with this because reduction in total fat seemed to be a simpler message than advice to replace saturated fat with polyunsaturated fat, which was based primarily on controlled feeding studies.
Because protein intake does not vary greatly across diets, a reduction in total fat implies an increase in carbohydrates, which were widely promoted as healthy choices without any direct evidence. This was epitomized by the original USDA Food Guide Pyramid of 1992, which suggested up to 11 servings a day of grains, along with including potatoes as a vegetable. Because about 80% of the carbohydrates in the US food supply are from refined starch, sugar, and potatoes, the result was that intake of these sources of calories increased by roughly 300 calories per day and the percent of energy from fat decreased from about 40% of energy to about 33% of energy.9 This corresponded with the huge increase in rates of obesity and diabetes in the United States; as discussed later, this dietary advice was probably contributory. Further, controlled feeding studies conducted in the 1980s and 1990s documented that high intakes of carbohydrate, compared to unsaturated fats or protein, increase triglycerides and blood pressure and reduce HDL cholesterol, and LDL particle size, which would all predict higher rates of heart disease.5,10
Randomized Trials and Cohort Studies of Diet and Disease
Motivated by the international correlations between dietary factors and rates of CHD and cancers, more detailed studies were needed to understand the specific dietary factors that might be involved. Ideally, hypotheses about diet and health outcomes would be answered by randomized trials in which many thousands of people are assigned to high intake of a dietary factor and others to low intake. However, such studies are extremely costly, and even when conducted they have often been uninformative because of low adherence to the assigned diets. The Women’s Health Initiative low-fat diet trial among 48,000 women provides the most recent example;11 at no time during the 7 years of follow-up did plasma levels of HDL cholesterol and triglycerides differ between the low-fat and usual diet groups.12,13 Because these lipids are known to change with low-fat diets, the trial was unfortunately uninformative. Trials of cancer prevention are even more challenging, because the effects of dietary change may not be seen until decades later. For these reasons, prospective cohort studies in which many thousands of persons are followed for many years are likely to be the most useful source of evidence about the relation between diet and major diseases. The Nurses’ Health Study (NHS) was among the early cohort investigations of diet and health, and the only one of these to repeatedly update dietary information over time, beginning in 1980. Especially when findings for health outcomes, such as coronary heart disease, are consistent with controlled feeding studies with intermediate outcomes, such as blood lipids or blood pressure, reproducible prospective cohort studies can provide sufficient support for causality to make dietary recommendations and nutrition policy.14
Specific Types of Dietary Fats and Heart Disease
Over the last 60 years, findings from controlled feeding studies of blood lipids, prospective cohort studies, and some randomized trials have provided strong evidence that replacing saturated fat with polyunsaturated fat will reduce coronary heart disease.15,16 Based on this concept, consumption of polyunsaturated fat, until recently mainly as soybean oil and corn oil, approximately doubled in the United States since the 1960s and is almost certainly one of the major contributors to the large decline in coronary heart disease mortality, by more than 60%.
Considerable confusion was caused by a recent meta-analysis by Chowdhury et al17 concluding that replacement of saturated fat with polyunsaturated fat for reduction of CHD was not supported by evidence. This meta-analysis was deeply flawed in many ways as described in the correspondence, but most fundamentally because it ignored the central concept in nutrition of isocaloric replacement; if we decrease one important source of calories, the effect will depend on what replaces it. In a meta-analysis of published studies, it was not possible to specify the replacement calories because individual studies have not reported their findings in a consistent way. For this reason, Jakobsen et al had earlier conducted an analysis using the original individual data from cohort studies of dietary fats and coronary heart disease.18 In that analysis, confirming an earlier, detailed analysis of NHS data,19 higher intake of saturated fat was not significantly associated with risk of coronary heart disease when compared with a similar number of calories from carbohydrate. This is consistent with the metabolic evidence that this replacement reduces both LDL cholesterol and HDL cholesterol and increases triglyceride levels.20 However, higher intake of saturated fat was associated with greater risk of coronary heart disease when compared with a similar number of calories from polyunsaturated fat, again consistent with well-documented changes in blood lipids. A similar relationship has been seen with total mortality.16 In controlled feeding studies, monounsaturated fat has effects on blood lipids that are similar to polyunsaturated fat,20 but cohort studies are complicated by the fact that in Western diets most monounsaturated fats come from meat, dairy products, and partially hydrogenated vegetable oils. In a detailed analysis within the NHS, one of the few studies with details on types of fat, monounsaturated fat was also inversely associated with risk of CHD when compared isocalorically with carbohydrate or saturated fat.19 The recent PREDIMED randomized trial,21 in which the addition of nuts or olive oil to a traditional Mediterranean diet reduced the incidence of cardiovascular disease compared to a low-fat diet, is consistent with benefits of monounsaturated fat, although this could not be completely disentangled from other components of the overall Mediterranean diet.
The picture of dietary fat and cardiovascular disease was complicated by recognition of trans-fatty acids, produced by the partial hydrogenation of vegetable oils, as an important element of our food supply. By the mid-1900s the large majority of vegetable oils in the United States were partially hydrogenated because this extended shelf life and produced solid fats that could mimic lard and butter, the traditional fats of northern European diets. However, in the 1970s, we and others became concerned that partial hydrogenation might have serious adverse effects because the oils being hydrogenated were primarily the essential fatty polyunsaturated fatty acids, linoleic and alpha-linolenic acid (ALA), that are critical to the structure of every cell membrane and the precursor molecules for the prostaglandins and other critical molecules. Thus, altering the double bonds of these essential fatty acids from their natural cis to the trans position by partial hydrogenation would almost surely change their functions in ways that were unpredictable and potentially harmful. This issue was of particular concern because intake of trans fat in the form of margarine and vegetable shortening was being strongly promoted for prevention of coronary heart disease because these products were believed to be superior to butter and lard due to a lower saturated fat content. The findings from multiple large cohort studies showing elevated risks of CHD with higher trans fat intake and controlled feeding studies showing that trans fats uniquely increase LDL and reduce HDL cholesterol along with other adverse effects22 led to compelling conclusions that intake should be as low as possible. Through a combination of education, product reformulation, and banning in restaurants in many cities and states, consumption has greatly decreased,23 LDL has decreased and HDL has increased in children and adults,24,25 and the FDA has effectively banned trans fats as of 2018.
Evidence on types of fat and heart disease is summarized in Figure 10.1. If saturated fat were replaced by trans fat, risk of heart disease would increase. However, if saturated fat were replaced by unsaturated fat, especially polyunsaturated fat, risk would decrease. In practice, this means replacing dairy fats and fats from red meat with liquid vegetable oils wherever possible.26 If saturated fat is replaced with the average carbohydrate in the American diet, there would be little effect on risk of heart disease. However, the average carbohydrate in the US diet, about half of total calories, is approximately 80% refined starch, sugar, or potatoes; as discussed later, the quality of carbohydrate also has important impacts on risk of cardiovascular disease, and if saturated fat is replaced by whole grains, risk of heart disease would be lower.26,27 Given the mix of types of fat and carbohydrate in the United States, total fat intake has not been associated with risks of heart disease;19 the type of fat and carbohydrate make a major difference.
Although polyunsaturated fat was considered as a homogeneous class of fatty acids in early diet–heart studies, much evidence indicates that the two main families, the N-6 and N-3 fatty acids, have many unique and important roles; both are considered essential fatty acids. A widespread belief has emerged, based on a simplistic interpretation of metabolic pathways, that N-6 fatty acids are proinflammatory and increase risks of heart disease, mental illness, and other conditions, and that the ratio of N-3 to N-6 fatty acids is critical. In fact, both N-6 and N-3 fatty acids are essential and beneficial; thus the ratio is irrelevant;28,29 in fact, it is possible to have a “perfect” ratio with deficiencies of both types of fatty acids. Linoleic acid, the primary dietary N-6 fatty acid, is a precursor of proinflammatory eicosanoids, but this conversion is highly regulated; and in controlled feeding studies higher intake of dietary linoleic acid does not increase proinflammatory eicosanoids or indicators of inflammation such as C-reactive protein (CRP).30 Indeed, in about half of the studies, increasing intake of linoleic acids was associated with lower levels of inflammatory factors, probably because it down-regulates NF-kappa B and can increase insulin sensitivity. Most importantly, in the epidemiologic studies and randomized trials described earlier, linoleic acid composed the large majority of the polyunsaturated fat, which is related to lower risk of coronary heart disease. In a recent reanalysis of the Minnesota Heart Study,31 the role of linoleic acid in the prevention of heart disease was questioned, but the average follow-up was less than 2 years and by the 3rd year approximately 85% of the participants had been lost to follow-up, rendering any conclusions uninterpretable. The relation between linoleic acid and risk of CHD in cohort studies was recently summarized;29 as shown in Figure 10.1, for total polyunsaturated fat intake, compared to carbohydrate or saturated fat, intake of linoleic acid was related to lower risk of CHD. Intake of total polyunsaturated fat and linoleic acid have also been examined extensively in relation to risk of breast and other cancers, diabetes, and many other conditions without evidence of harm.32 In a recent large analysis of over 175,000 men and women, higher intake of N-6 polyunsaturated fat intake was strongly related to lower total mortality.16 Thus, within the range consumed in usual human diets, up to about 9% of calories in the US diet, N-6 fatty acids are both essential and beneficial.
Like N-6 fatty acids, N-3 fatty acids play many essential structural and functional roles, and they have also been hypothesized to have a unique role in preventing cardiac arrhythmias and thus sudden cardiac death.33 In human diets, fish oils are the major source of extra-long-chain N-3 fatty acids, but ALA, primarily from plants and especially soy and canola oils, is quantitatively the primary source of total N-3 fatty acids. In studies using dietary intake or levels in red blood cells, docosohexaenoic acid (DHA) and eicosapentaenoic acid (EPA) have been inversely associated with fatal coronary heart disease in cohort studies,15 and reductions in risk of CHD have been seen in some randomized trials.15 In several recent randomized trials, supplements of EPA/DHA have not reduced cardiac deaths, raising doubts about their benefits. However, randomized trials of dietary supplements are not like trials of new drugs in which the placebo group has no intake of the agent being tested; in the EPA/DHA trials everyone had some intake of N-3 fatty acids at baseline, and if intake was already near optimal for many participants, supplements would have no apparent benefit. This does not exclude a potential benefit among a significant minority in the population or an important overall benefit in other populations, which include major parts of the world with much lower intakes.34 Studies of ALA and cardiovascular diseases are fewer, but a benefit appears likely where fish intake, and thus EPA/DHA intake, is low.35 Although the optimal intakes of EPA/DHA and ALA remain to be determined, the AHA recommendation to eat fish at least twice per week is reasonable, and using a variety of plant oils, especially including canola or soybean oil, will provide a source of ALA. Walnuts and flaxseed are other potential sources.
Because the presumed relation of total fat intake to risk of breast cancer, based primarily on international correlations, was a key pillar for national dietary recommendations, this was examined in detail at repeated intervals in the Nurses’ Health Study; at no time was there any suggestion of a positive relationship.36 This lack of any important relation has been confirmed in a summary of cohort studies; except for a possible inverse association with monounsaturated fat, specific types of fat have also not been associated importantly with risk of breast cancer.32,37 The findings for total fat are also consistent with randomized trials,11,38 although the WHI trial findings are uninformative, as noted earlier.13 A similar lack of increased risk with total fat has also been seen for cancers of the colon and rectum, and for prostate cancer; indeed, for no cancer examined has total fat intake been found to be a risk factor.
For type 2 diabetes, adiposity is a powerful risk factor, but total dietary fat is not. As for coronary heart disease, trans fat intake is importantly related to increased risk,39,40 and polyunsaturated fat or the ratio of polyunsaturated to saturated fat is related to lower risk.
Dietary Fat and Body Weight
Perhaps the most deeply embedded belief about the benefits of low-fat/high-carbohydrate diets has been that they are effective in preventing weight gain and treating overweight. This was supported by some early randomized trials,41 but most of these lasted only a few months and in most trials the low-fat group received intensive intervention and the control group received none. With additional, longer studies, it became apparent that the weight loss in groups assigned to low-fat diets was usually transient, with a nadir at about 6 months, and that the intensity of intervention itself contributes to weight loss. In the most comprehensive meta-analysis of randomized trials with data on dietary fat and body weight that lasted for 1 year or longer, among trials with equal intensity of intervention, persons assigned to low-fat diets lost less weight than the control groups with higher fat intake.42 Although, low-fat/high-carbohydrate diets are not generally effective for weight control, other characteristics of diet do have important influences. In a comprehensive analysis of specific foods among over 100,000 men and women, potatoes, red meat, soda, and fruit juice were related to more weight gain, and whole grains, fruits, vegetables, and yogurt were related to less weight gain.43 The foods associated with less weight gain are consistent with a Mediterranean dietary pattern, which was evaluated in a compelling study of diet and weight control, conducted by Shai et al.44 In that study, employees at a worksite were randomized to a low-fat diet, a high-fat diet, and a Mediterranean diet with fat content just slightly lower than the high-fat diet. By 6 months, weight loss was similar on all diets, but by 2 years, the low-fat group had regained most of their weight. Most notably, after an additional 4 years without intervention, the low-fat group had regained all their weight, the Mediterranean diet group had maintained most of their weight loss and had improved metabolic variables, and the high-fat group was intermediate.45 The sustained weight loss of the Mediterranean group suggests that the participants had internalized this eating pattern, which is the ultimate goal of an intervention.
Notably, in weight loss trials the primary reported results are average changes in weight, but what has been remarkably consistent is the wide variation in response to changes in dietary fat. On all diets, some persons lose large amounts of weight, but others lose nothing or gain weight. The explanation for this heterogeneity remains unclear but could be related to biological factors or social factors, such as family support. However, this does suggest a role for individual experimentation to find an eating pattern that is effective for them in long-term weight control; an essential criterion is that this must be sufficiently varied, satisfying, and enjoyable so that it can be maintained for years. This likely explains the success of the Mediterranean diet in the study by Shai et al. Of course, a long-term diet must also be compatible with overall good health, which must be evaluated in much larger and longer-term studies.
In summary, despite vast efforts to document a benefit of low-fat diets for prevention of cancer, cardiovascular disease, weight control, and many other health outcomes, such benefits have not emerged. For this reason, the 2015 US Dietary Guidelines removed the earlier upper limits on the percentage of energy from fat, which will allow a wider range of healthy diets.
Until recently, intakes of total and specific types of carbohydrates have received much less attention than dietary fat, which is surprising because they are the major source of calories. Because protein intake varies relatively little among persons, the relation of total carbohydrate to health outcomes tends to be the reciprocal of total fat intake; and as for total fat intake, there is insufficient evidence to set limits for carbohydrate intake. Indeed, there is no nutritional requirement for carbohydrate. Similar to fat, the type of carbohydrate has important health implications; unfortunately, there is no single, simple measure of carbohydrate quality, but it can be characterized by its degree of refinement (such as whole grain or fiber content), its glycemic index, and whether in solution or solid form.
The refining of grains has major impacts on their nutritional value because the large majority of fiber, minerals, and vitamins are removed. The remaining starch is simply a chain of glucose molecules with no nutritional value beyond a source of energy and, when broken down into glucose, multiple potential adverse metabolic effects. In contrast, consumption of cereal fiber or whole grains has been consistently associated with lower risks of cardiovascular disease,46 diabetes,47 and digestive disorders; evidence for cancer is less consistent.32,48 As they travel together in whole grains, the contributions of cereal fiber, minerals, and vitamins are difficult to distinguish.
The glycemic index of a carbohydrate, assessed as the incremental increase in blood glucose after fixed amount of carbohydrate is consumed, is influenced by many factors, but the pulverizing of grain into fine flour is particularly important, as this increases the rate of enzymatic conversion of starch to glucose. Because the effect of a food on blood glucose levels will depend on both the amount of carbohydrate and the glycemic index of this carbohydrate, we have developed the glycemic load, which is calculated by multiplying the glycemic index by the amount of the carbohydrate in a food or diet. Along with refined grains, potatoes are also a major source of glycemic load in Western diets because they have large amounts of carbohydrate with a high glycemic index. The concept of glycemic index has been controversial, but in our cohort studies and a meta-analysis, higher glycemic index is importantly associated with higher risks of type 2 diabetes,49 and glycemic load strongly predicts risk of CHD.50 This evidence is supported by a large randomized trial of a glucosidase inhibitor,51 which pharmacologically converts high glycemic index foods to low glycemic index foods within the gastrointestinal tract; incidence of diabetes and cardiovascular disease was reduced.
Liquid carbohydrates, mainly as soda and other sugar-sweetened beverages (SSBs), are particularly problematic, because very large amounts of sugar can be consumed rapidly with minimal induction of satiety. A large body of evidence has documented adverse effects of SSBs on weight gain and risks of diabetes and coronary heart disease.52 Some have suggested that fructose, half of the sucrose molecule and only slightly more than half of high fructose corn syrup, is substantially more harmful than glucose. Although fructose is metabolized differently than glucose, they both have harmful metabolic effects, and keeping intake of added sugar low, especially in the form of soda and other sugar-sweetened beverages, is a high priority.
Few foods contain protein as the main form of calories, so the health impacts of dietary protein are best considered as part of the food in which it is packaged, which often includes a substantial amount of fat, which varies greatly in type. Although current US dietary guidelines make little distinction between sources of proteins,52 the health consequences differ greatly. Red meat stands out as being related to higher risks of CHD, stroke, type 2 diabetes, and some cancers.53,54,55 In analyses examining alternatives to red meat, risks of these endpoints are lower with intakes of poultry, fish, low-fat dairy foods, beans, and nuts (see http://www.iarc.fr/en/media-centre/iarcnews/pdf/Monographs-Q&A_Vol114.pdf). Some of these differences are due to a higher proportion of unsaturated fatty acids in most of these alternatives, but the high amounts of heme iron and cholesterol in red meat and other microconstituents may also contribute to the elevated risks. There is no evidence that eating leaner meats mitigates the adverse effects; for example, heme iron and cholesterol are predominately in the lean part of meat.
Apart from the direct health effects of red meat, the impacts of high red meat consumption on environmental sustainability, essential to food security, are a concern. Whether expressed in relation to caloric or protein content, red meat has by far the greatest contribution to greenhouse gas production compared to any other food, and also has disproportionally large water, energy, and land footprints (see http://www.ewg.org/meateatersguide/a-meat-eaters-guide-to-climate-change-health-what-you-eat-matters/climate-and-environmental-impacts/).
Dairy and Vitamin D
High consumption of dairy foods, typically three servings a day, has been promoted, primarily based on expected reductions in osteoporosis and fractures because of its high calcium content. However, the high recommended daily allowances (RDAs) for calcium intake in the United States (e.g., 1,200 mg/day for women over 50 years) are based on studies lasting only 2 weeks, which is far too short to assess long-term calcium needs; the World Health Organization (WHO) has concluded that 500 mg/day is adequate. Further, in a meta-analysis of prospective studies, high intakes of calcium and milk were not associated with lower risk of fractures.56 Dairy food consumption is not importantly related to weight gain57 or risk of type 2 diabetes,58 but is related to higher risk of prostate cancer and lower risk of colorectal cancer.32 Given these mixed findings, keeping dairy consumption in the range of 0 to 2 servings a day seems sensible; if few or no dairy foods are consumed, taking 500 mg of calcium per day as a safety net is reasonable.
Vitamin D is essential for bone health. Fortification of milk has been an effective public health strategy that almost eliminated rickets in children, and supplements of vitamin D in the range of 700 to 800 mg/day have reduced fractures in older adults.59 Considerable evidence suggests that low vitamin D levels may be contributing to risks of other diseases including colorectal cancer and multiple sclerosis.60,61 The natural source of vitamin D is solar exposure, and fish is the only food that contains an appreciable amount of vitamin D. Although vitamin D supplementation remains controversial, taking 1,000 IU per day has been deemed safe62 and seems reasonable unless someone is routinely exposed to substantial outdoor sunlight. Some people may require substantially more to minimize disease risk, but optimal intakes are not yet clear.
Fruits and Vegetables
Fruits and vegetables contain a wide range of essential nutrients and many other constituents that are hypothesized to reduce risks of many diseases. The strongest evidence is that these foods reduce blood pressure63 and risk of cardiovascular disease. At least some of the benefit is mediated by their content of potassium64 and probably folate. These foods have much less impact on cancer risk than had earlier been believed, but reduction in risk of ER-negative breast cancer is likely.65 The composition of fruits and vegetables varies widely, and it is thus unlikely that they have similar health effects. Potatoes should not be considered a vegetable, but rather a form of starch along with corn.66,67,68 In the United States, many consume few green leafy vegetables, which are important to include, and berries appear to be particularly beneficial for prevention of diabetes risk.69 Although more remains to be learned about specific foods, consuming at least five servings a day of nonstarchy fruits and vegetables is an important goal.
Alcohol, Coffee, and Tea
Heavy alcohol consumption has many adverse health effects and should be avoided; the balance of risks and benefits for low and moderate consumption has been controversial. Very briefly, evidence is strong that moderate alcohol consumption reduces risk of coronary heart disease, and also lower risks of total mortality in middle-aged adults.70 This benefit appears to be due primarily to ethanol per se rather than other constituents of these beverages. The major counterbalancing health effect at the level of 1–2 drinks per day is an increased risk of breast cancer.71,72,73,74,75,76
Early concerns about possible increases in risks of cancer and cardiovascular disease with coffee consumption have been erased by a large literature showing no increases in risk for any type of cancer, substantial decreases in risk of diabetes, and a likely modestly lower risk of total mortality.77,78
Few topics are as controversial as nutritional supplements. If everyone were consuming an ideal diet, a benefit of supplements would likely be small, except for vitamin D, because foods are a poor source. However, fewer than 5% meet the US dietary guidelines recommendations, and we found that none of low-income Americans in a national survey came close to consuming an optimal diet.79 The fact that multiple vitamins dramatically reduce risk of birth defects and that vitamin D fortification virtually eliminated rickets, demonstrates that diets are not optimal. Some of the controversy over the value of vitamin supplements stems from “negative” randomized trials that were of naively short duration for conditions that develop over decades and/or that likely include many participants who already have adequate intake.80 For example, the Physician’s Health Study was the only trial that lasted for more than a decade, and in that study beta-carotene improved cognitive function relative to placebo81 and a multiple vitamin reduced total cancer incidence by 8%, which is remarkable because this was an unusually well nourished population.82 Given available evidence, use of an RDA level multiple vitamin with 800 to 1,000 IU of vitamin D is reasonable for most people. For women of reproductive age, who should already be taking a multiple vitamin with folate for prevention of birth defects, inclusion of an RDA level of iron is also reasonable because approximately 15% of women in the group are iron deficient.
The Overall Impact of a Healthy Diet: Dietary Patterns and Indices
As described earlier, and as shown in Figure 10.2, the key elements of health eating are:
• An abundance of fruits and vegetables, not including potatoes.
• Healthy fats from plant sources, including virtually all liquid vegetable oils.
• Protein from predominantly plant sources, with optionally moderate amounts of fish, poultry, and dairy foods. Red meat to be occasional at most.
• Whole grains, with limited amounts of sugar and refined grains.
One way of evaluating the healthfulness of an overall diet is to examine an overall dietary pattern in relation to health outcomes. Patterns, such as the traditional Mediterranean dietary pattern,83 can be defined by using a scoring system, giving points for elements included in a diet or subtracting points for elements not in that diet. Within many countries, adherence to the Mediterranean dietary pattern has been associated with lower risk of cardiovascular disease, diabetes, cognitive decline, and premature mortality. Most of these benefits have been confirmed in the PREDIMED randomized trial conducted in Spain;21 and in a randomized trial among post-myocardial-infarction patients, a Mediterranean diet reduced recurrence or cardiovascular death by 70%.84 Another approach has been to define a dietary score or index based on the overall literature regarding the elements of a healthy diet; for example, we created the Alternative Healthy Eating Index (AHEI) because the 2000 USDA Healthy Eating Index (HEI) failed to predict health outcome. With later important changes to the HEI, both indices strongly predict better health outcomes.85,86 Collectively, these analyses document the importance of a healthy diet across many different outcomes and in many different populations around the world. Our understanding of the elements of a healthy diet allows great flexibility using different foods, flavors, and culinary traditions.
The potential for reduction of major diseases by diet and lifestyle has been examined using data from large cohort studies. Using data from the NHS, the combination of not smoking, being moderately active, eating a healthy diet (defined by a simple five-element score), maintaining a healthy weight, and moderate alcohol consumption was estimated to reduce incidence of coronary heart disease by 82%;87 the contributions of better diet, physical activity, avoidance of overweight, and not smoking were similar. Similar findings for younger women were recently reported from the Nurses’ Health Study II.88 Using the same approach, 92% of type 2 diabetes and 70% of stroke were estimated to be preventable by a similar set of diet and lifestyle factors.89
Encouragingly, through education and policies, dietary quality is gradually improving in the United States, although the room for further improvement is huge. Using the AHEI to score dietary data collected by the NHANES from 1999 through 2012, the score increased from approximately 40 to 50 on a scale of 110.23 The greatest improvement was due to an approximately 80% reduction in intake of trans fat, and decline in soda consumption of about 25% also contributed importantly. Modest increases were seen for whole grains, fruits, and polyunsaturated fat. Only sodium intake trended in an adverse direction. Although the average changes were encouraging, most of the improvements in dietary quality were among those with higher income and education and among those with healthy weights, thus identifying areas where greater efforts are needed. In this same report, it was estimated that more than 1 million premature deaths were prevented by improvements in diet quality from 1999 through 2012, and that these improvements would reduce incidence of type 2 diabetes by about 12%. Encouragingly, the CDC has recently reported that incidence of type 2 diabetes had decreased by 20% in the United States after years of continuous increases (see http://www.cdc.gov/diabetes/statistics/incidence/fig2.htm).
Putting Knowledge into Practice
Unfortunately, medical schools and postgraduate programs have rarely included more than minimal information about diet and health and how to incorporate this into practice.2 Although we have learned much about the relation of diet to health over the last several decades, evidence about the most effective ways to integrate this into preventive health care is still rudimentary. This will include basic skills in motivational interviewing that will not be covered here, and a variety of approaches depending on the context and individual patients. Our longer experience in control of smoking provides some guidance, and management of weight is discussed elsewhere in this book (see chapter 18). Some general considerations include:
1. Most importantly, healthcare providers should practice healthy dietary practices themselves. Doing so will make providers more knowledgeable about healthy diets and the benefits they provide as well barriers to healthy eating and ways to overcome these obstacles.
2. Determine a patient’s BMI and weight change since age 20 and over the past year or two. Although these are simple and inexpensive to assess, they are often not collected or are ignored, even though they strongly predict many health outcomes and total mortality.90 Assessment of waist circumference is also desirable, especially as people pass through midlife. Weight gain is important, because increases of more than 5–10 pounds predict adverse outcomes, even though BMI can remain below the standard cutoff of 25 kg/m2. Also, this indicates that a person is on a trajectory to gain more weight unless changes are made in diet or physical activity.
3. Assess a patient’s diet, even if crudely. If nothing more is possible, add soda consumption to the routine assessment of tobacco and alcohol use. Usual diet can be assessed inexpensively by self-administered food frequency questionnaires used for research purposes; these provide a wealth of information about a person’s overall diet and the individual constituents of that diet;91 as noted above, this information predicts many health outcomes and premature death. Simplified screening questionnaires can capture much of the information provided by more complex dietary assessment.92 Neither of these approaches has been routinely incorporated into clinical practice, leaving healthcare providers without information about one of the most important determinants of their patient’s future and without an informed basis to provide individual counseling. More research is clearly needed on the incorporation of dietary assessment in preventive healthcare.
4. Develop a menu of options for weight control and improvement in diet quality. Some patients are highly motivated and simply need scientifically based evidence to make changes in their diets that can have major benefits. To fill this need, I have written Eat, Drink and Be Healthy93 and, with Molly Katzen, Eat, Drink, and Weigh Less.94 Others will benefit from additional individual counseling or group support. In a recent randomized trial within a clinical setting, impressive weight loss was seen with both individual counseling and a less expensive intervention that used computer-generated follow-up reminders.95
5. Take advantage of teachable moments. Although we would ideally prevent a large proportion of intermediate risk factors—like hypertension, hyperlipidemia, colon adenomas, or metabolic syndrome—by healthy diet and lifestyle, the large majority of people who develop these conditions do not have an optimal diet. The diagnosis of these conditions provides an extra reason to review diet and other lifestyle factors, and potentially additional motivation for the patient. Similarly, the diagnosis of diabetes, cardiovascular disease, or other diet-related condition in a family member can provide a good opportunity to review risk factors, as these people may well be sharing a common diet. Unfortunately, these diagnoses typically are lost opportunities. The reporting of the SPRINT study, showing that a lower target for systolic blood pressure reduced risk of stroke,96 provides an example. In that study the average BMI of participants was 30 kg/m2, meaning the large majority were overweight, which is the most important cause of hypertension (see http://www8.nationalacademies.org/onpinews/newsitem.aspx?RecordID=12819). Yet the published paper, editorial, and news reports only described the need for further medication (which added to side effects), and did not consider why patients had hypertension (overweight is the primary cause)64,97 or that blood pressure can be lowered by well-documented changes in diet.97
6. Engage directly in enhancement of diets. The role that healthcare providers play in helping patients improve their diet can vary greatly depending on their knowledge, financial and time constraints, and wishes to be deeply involved in this process. At a minimum, a healthcare provider should inform patients of their concern about overweight, a creeping weight gain, excessive soda consumption, or a condition that might be modified by diet. In the case of smoking, just a simple message of concern has a measurable impact on smoking cessation. The next level of engagement would be referral to more information, a group program such as Weight Watchers, a comprehensive lifestyle intervention program,98 or individual counseling by health coach or dietitian. Some healthcare providers choose to become deeply involved in dietary assessment, nutrition counseling, or hands-on cooking demonstrations, which can be on an individual or group basis.
7. Consider expanding your influence. While healthcare providers can do much to help their patients improve their diets, many powerful factors influence what we eat, including our family, workplace, schools, food availability in our community, marketing, economic factors, and local, national, and international policies. In Thinfluence, Dr. Malissa Wood and I describe ways that individuals can modify or circumvent these external factors to improve their own diets and the well-being of others around them. Healthcare providers can play a special role because of our knowledge and credibility in health-related matters.99 Hospitals, where many of us work, should be providing the very best example of healthy eating for both patients and staff, but have unfortunately often provided just the opposite. Physicians should be leading the effort to improve this environment, and in places important changes are being made.100,101 Many other opportunities for leadership exist in our communities, schools, worksites, and other institutions.
8. Avoid nihilism about dietary change. Caring for patients who need to change their diet, or who need to stop smoking, can sometimes be discouraging, and relapses to previous habits are frequent. While some persons make transformational and permanent changes, changes frequently happen slowly, and often after many failed attempts to improve. This is partly because of the many powerful influences on behavior that are beyond our direct control. The best-published smoking interventions have success rates on the order of only 10% to 15% after 1 year, which is hardly detectable by a practicing physician. Yet, over a period of 40 years, smoking rates in men have decreased from about 60% to less than 20%. It is hard to know the specific role that healthcare providers have played in this achievement, but almost certainly their contribution to education, awareness, and motivation has been important. As noted above, as a country we are making progress in improving dietary quality, and some individuals have experienced this directly and profoundly. However, far more needs to be done; we have the knowledge of what should be done, and healthcare providers can do much for their individual patients and our society to make this a reality.
Optimal diets can play a major role in prevention of the major causes of morbidity and mortality in Western populations; this has been documented by a wealth of data from controlled feeding studies with risk factors as endpoints, long-term prospective studies, and randomized trials. The key elements of a healthy diet are unsaturated sources of fat from plants and plant oils; whole grains; beans, nuts, soy, fish, and poultry as major protein sources; and an abundance of fruits and vegetables. Intake of trans fat, processed meats, and sugar-sweetened beverages should be avoided, and consumption of red meat, dairy fat, refined grains, and fruit juices should be limited. The traditional Mediterranean diet incorporates these elements, and the health benefits of this dietary pattern have been proven. Other combinations of foods and flavors that are based on these elements can be similarly healthy. Physicians and healthcare providers can play an important role in guiding their patients toward healthy eating patterns, and they can also provide leadership in developing a healthier food environment for all.
1. Lozano R, Naghavi M, Foreman K, et al. Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012;380:2095–128.Find this resource:
2. Devries S, Dalen JE, Eisenberg DM, et al. A deficiency of nutrition education in medical training. Am J Med 2014;127:804–806.Find this resource:
3. Keys A.Seven Countries: A Multivariate Analysis of Death and Coronary Heart Disease. Cambridge, MA: Harvard University Press; 1980.Find this resource:
4. Kato H, Tillotson J, Nichaman MZ, Rhoads GG, Hamilton HB. Epidemiologic studies of coronary heart disease and stroke in Japanese men living in Japan, Hawaii and California. Am J Epidemiol 1973;97:372–385.Find this resource:
5. Mensink RP, Katan MB. Effect of dietary fatty acids on serum lipids and lipoproteins: a meta-analysis of 27 trials. Arterioscler Thrombosis 1992;12:911–919.Find this resource:
6. Armstrong B, Doll R. Environmental factors and cancer incidence and mortality in different countries, with special reference to dietary practices. Int J Cancer 1975;15:617–631.Find this resource:
7. Buell P. Changing incidence of breast cancer in Japanese-American women. J Natl Cancer Inst 1973;51:1479–1483.Find this resource:
8. National Research Council, Committee on Diet and Health. Diet and Health: Implications for Reducing Chronic Disease Risk. Washington, DC: National Academy Press; 1989.Find this resource:
9. Centers for Disease Control and Prevention, National Center for Health Statistics. Daily dietary fat and total food-energy intakes: Third National Health and Nutrition Examination Survey, Phase 1, 1988–91. MMWR 1994;43(7):116–117.Find this resource:
10. Appel LJ, Sacks FM, Carey VJ, et al. Effects of protein, monounsaturated fat, and carbohydrate intake on blood pressure and serum lipids: results of the OmniHeart randomized trial. JAMA 2005;294:2455–2464.Find this resource:
11. Prentice RL, Caan B, Chlebowski RT, et al. Low-fat dietary pattern and risk of invasive breast cancer: the Women’s Health Initiative Randomized Controlled Dietary Modification Trial. JAMA 2006;295:629–642.Find this resource:
12. Howard BV, Curb JD, Eaton CB, et al. Low-fat dietary pattern and lipoprotein risk factors: the Women’s Health Initiative Dietary Modification Trial. Am J Clin Nutr 2010;2010;91(4):860–874.Find this resource:
13. Willett WC. The WHI joins MRFIT: a revealing look beneath the covers. Am J Clin Nutr 2010;91(4):829–830.Find this resource:
14. Willett WC. Policy applications. In: Willett WC, ed. Nutritional Epidemiology. 3rd ed. New York, NY: Oxford University Press; 2013:357–379.Find this resource:
15. Willett WC, Stampfer MJ. Diet and coronary heart disease. In: Willett WC, ed. Nutritional Epidemiology. 3rd ed. New York, NY: Oxford University Press; 2013:426–467.Find this resource:
16. Wang DD, Li Y, Chiuve SE, et al. Association of specific dietary fats with total and cause-specific mortality. JAMA Intern Med 2016 Aug 1;176(8):1134–1145.Find this resource:
17. Chowdhury R, Warnakula S, Kunutsor S, et al. Association of dietary, circulating, and supplement fatty acids with coronary risk: a systematic review and meta-analysis. Ann Intern Med 2014;160:398–406.Find this resource:
18. Jakobsen MU, O’Reilly EJ, Heitmann BL, et al. Major types of dietary fat and risk of coronary heart disease: a pooled analysis of 11 cohort studies. Am J Clin Nutr 2009;89:1425–1432.Find this resource:
19. Hu F, Stampfer MJ, Manson JE, et al. Dietary fat intake and the risk of coronary heart disease in women. N Engl J Med 1997;337:1491–1499.Find this resource:
20. Mensink RP, Zock PL, Kester AD, Katan MB. Effects of dietary fatty acids and carbohydrates on the ratio of serum total to HDL cholesterol and on serum lipids and apolipoproteins: a meta-analysis of 60 controlled trials. Am J Clin Nutr 2003;77:1146–1155.Find this resource:
21. Estruch R, Ros E, Salas-Salvado J, et al. Primary prevention of cardiovascular disease with a Mediterranean diet. N Engl J Med 2013;368:1279–1290.Find this resource:
22. Willett WC, Stampfer MJ, Manson JE, et al. Intake of trans fatty acids and risk of coronary heart disease among women. Lancet 1993;341:581–585.Find this resource:
23. Wang DD, Li Y, Chiuve SE, Hu FB, Willett WC. Improvements in US diet helped reduce disease burden and lower premature deaths, 1999–2012; overall diet remains poor. Health Aff (Millwood) 2015;34:1916–1922.Find this resource:
24. Kit BK, Carroll MD, Lacher DA, Sorlie PD, DeJesus JM, Ogden C. Trends in serum lipids among US youths aged 6 to 19 years, 1988–2010. JAMA 2012;308:591–600.Find this resource:
25. Carroll MD, Kit BK, Lacher DA, Shero ST, Mussolino ME. Trends in lipids and lipoproteins in US adults, 1988–2010. JAMA 2012;308:1545–1554.Find this resource:
26. Chen M, Li Y, Sun Q, et al. Dairy fat and risk of cardiovascular disease in 3 cohorts of US adults. Am J Clin Nutr 2016 Nov;104(5):1209–1217.Find this resource:
27. Li Y, Hruby A, Bernstein AM, et al. Saturated fats compared with unsaturated fats and sources of carbohydrates in relation to risk of coronary heart disease: a prospective cohort study. J Am Coll Cardiol 2015 Oct 6;66(14):1538–1548.Find this resource:
28. Hu FB, Stampfer MJ, Manson JE, et al. Dietary intake of α-linolenic acid and risk of fatal ischemic heart disease among women. Am J Clin Nutr 1999;69:890–897.Find this resource:
29. Farvid MS, Ding M, Pan A, et al. Dietary linoleic acid and risk of coronary heart disease: a systematic review and meta-analysis of prospective cohort studies. Circulation 2014;130:1568–1578.Find this resource:
30. Fritsche KL. Too much linoleic acid promotes inflammation—doesn’t it? Prostaglandins Leukot Essent Fatty Acids 2008;79:173–175.Find this resource:
31. Ramsden CE, Zamora D, Majchrzak-Hong S, et al. Re-evaluation of the traditional diet-heart hypothesis: analysis of recovered data from Minnesota Coronary Experiment (1968–73). BMJ 2016;353:i1246.Find this resource:
32. World Cancer Research Fund/American Institute for Cancer Research. Food, Nutrition, Physical Activity, and the Prevention of Cancer: A Global Perspective. Washington, DC: AICR, 2007.Find this resource:
33. Bang HO, Dyerberg J, Sinclair HM. The composition of the Eskimo food in North Western Greenland. Am J Clin Nutr 1980;33:2657–2661.Find this resource:
34. Petrova S, Dimitrov P, Willett WC, Campos H. The global availability of n-3 fatty acids. Public Health Nutr 2011;14:1157–1164.Find this resource:
35. Mozaffarian D, Ascherio A, Hu F, et al. Interplay between different polyunsaturated fatty acids and risk of coronary heart disease in men. Circulation 2005;111:157–164.Find this resource:
36. Kim EH, Willett WC, Colditz GA, et al. Dietary fat and risk of postmenopausal breast cancer in a 20-year follow-up. Am J Epidemiol 2006;164(10):990–997.Find this resource:
37. Smith-Warner SA, Spiegelman D, Adami HO, et al. Types of dietary fat and breast cancer: a pooled analysis of cohort studies. Int J Cancer 2001;92:767–774.Find this resource:
38. Martin LJ, Li Q, Melnichouk O, et al. A randomized trial of dietary intervention for breast cancer prevention. Cancer Res 2011;71:123–133.Find this resource:
39. Hu FB, Manson JE, Stampfer MJ, et al. Diet, lifestyle, and the risk of type 2 diabetes mellitus in women. N Engl J Med 2001;345:790–797.Find this resource:
40. Kavanagh K, Jones KL, Sawyer J, et al. Trans fat diet induces abdominal obesity and changes in insulin sensitivity in monkeys. Obesity (Silver Spring) 2007;15:1675–1684.Find this resource:
41. Bray GA, Popkin BM. Dietary fat intake does affect obesity! Am J Clin Nutr 1998;68:1157–1173.Find this resource:
42. Tobias DK, Chen M, Manson JE, Ludwig DS, Willett W, Hu FB. Effect of low-fat diet interventions versus other diet interventions on long-term weight change in adults: a systematic review and meta-analysis. Lancet Diabetes Endocrinol 2015;3:968–979.Find this resource:
43. Mozaffarian D, Hao T, Rimm EB, Willett WC, Hu FB. Changes in diet and lifestyle and long-term weight gain in women and men. N Engl J Med 2011;364:2392–2404.Find this resource:
44. Shai I, Schwarzfuchs D, Henkin Y, et al. Weight loss with a low-carbohydrate, Mediterranean, or low-fat diet. N Engl J Med 2008;359:229–241.Find this resource:
45. Schwarzfuchs D, Golan R, Shai I. Four-year follow-up after two-year dietary interventions. N Engl J Med 2012;367:1373–1374.Find this resource:
46. Hu FB, Willett WC. Optimal diets for prevention of coronary heart disease. JAMA 2002;288:2569–2578.Find this resource:
47. Weickert MO, Pfeiffer AF. Metabolic effects of dietary fiber consumption and prevention of diabetes. J Nutr 2008;138:439–442.Find this resource:
48. Park Y, Hunter DJ, Spiegelman D, et al. Dietary fiber intake and risk of colorectal cancer: a pooled analysis of prospective cohort studies. JAMA 2005;294:2849–2857.Find this resource:
49. Bhupathiraju SN, Tobias DK, Malik VS, et al. Glycemic index, glycemic load, and risk of type 2 diabetes: results from 3 large US cohorts and an updated meta-analysis. Am J Clin Nutr 2014;100:218–232.Find this resource:
50. Augustin LS, Kendall CW, Jenkins DJ, et al. Glycemic index, glycemic load and glycemic response: an International Scientific Consensus Summit from the International Carbohydrate Quality Consortium (ICQC). Nutr Metab Cardiovasc Dis 2015;25:795–815.Find this resource:
51. Chiasson JL, Josse RG, Gomis R, et al. Acarbose for prevention of type 2 diabetes mellitus: the STOP-NIDDM randomised trial. Lancet 2002;359:2072–2077.Find this resource:
52. US Department of Agriculture, US Department of Health and Human Services. Scientific Report of the 2015 Dietary Guidelines Advisory Committee. Washington, DC: US Government Printing Offices; 2015.Find this resource:
53. Bernstein AM, Sun Q, Hu FB, Stampfer MJ, Manson JE, Willett WC. Major dietary protein sources and risk of coronary heart disease in women. Circulation 2010;122:876–883.Find this resource:
54. Farvid MS, Cho E, Chen WY, Eliassen AH, Willett WC. Adolescent meat intake and breast cancer risk. Int J Cancer 2015;136:1909–1920.Find this resource:
55. Pan A, Sun Q, Bernstein AM, Manson JE, Willett WC, Hu FB. Changes in red meat consumption and subsequent risk of type 2 diabetes mellitus: three cohorts of US men and women. JAMA Intern Med 2013;173:1328–1335.Find this resource:
56. Bischoff-Ferrari HA, Dawson-Hughes B, Baron JA, et al. Milk intake and risk of hip fracture in men and women: a meta-analysis of prospective cohort studies. J Bone Miner Res 2011;26:833–839.Find this resource:
57. Chen M, Pan A, Malik VS, Hu FB. Effects of dairy intake on body weight and fat: a meta-analysis of randomized controlled trials. Am J Clin Nutr 2012;96:735–747.Find this resource:
58. Chen M, Sun Q, Giovannucci E, et al. Dairy consumption and risk of type 2 diabetes: 3 cohorts of US adults and an updated meta-analysis. BMC Med 2014;12:215.Find this resource:
59. Bischoff-Ferrari HA, Willett WC, Wong JB, Giovannucci E, Dietrich T, Dawson-Hughes B. Fracture prevention with vitamin D supplementation: a meta-analysis of randomized controlled trials. JAMA 2005;293:2257–2264.Find this resource:
60. Dou R, Ng K, Giovannucci EL, Manson JE, Qian ZR, Ogino S. Vitamin D and colorectal cancer: molecular, epidemiological and clinical evidence. Br J Nutr 2016;115:1643–1660.Find this resource:
61. Simon KC, Munger KL, Ascherio A. Vitamin D and multiple sclerosis: epidemiology, immunology, and genetics. Curr Opin Neurol 2012;25:246–251.Find this resource:
62. Institute of Medicine. Dietary Reference Intakes for Calcium and Vitamin D. Washington, DC: National Academy of Sciences; 2010.Find this resource:
63. Moore TJ, Vollmer WM, Appel LJ, et al. Effect of dietary patterns on ambulatory blood pressure: results from the Dietary Approaches to Stop Hypertension (DASH) Trial. DASH Collaborative Research Group. Hypertension 1999;34:472–477.Find this resource:
64. Sacks FM, Willett WC, Smith A, Brown LE, Rosner B, Moore TJ. Effect on blood pressure of potassium, calcium, and magnesium in women with low habitual intake. Hypertension 1998;31:131–138.Find this resource:
66. Borgi L, Rimm EB, Willett WC, Forman JP. Potato intake and incidence of hypertension: results from three prospective US cohort studies. BMJ 2016;353:i2351.Find this resource:
67. Muraki I, Rimm EB, Willett WC, Manson JE, Hu FB, Sun Q. Potato consumption and risk of type 2 diabetes: results from three prospective cohort studies. Diabetes Care 2016;39:376–384.Find this resource:
68. Bertoia ML, Mukamal KJ, Cahill LE, et al. Changes in intake of fruits and vegetables and weight change in United States men and women followed for up to 24 years: analysis from three prospective cohort studies. PLoS Med 2015;12:e1001878.Find this resource:
69. Muraki I, Imamura F, Manson JE, et al. Fruit consumption and risk of type 2 diabetes: results from three prospective longitudinal cohort studies. BMJ 2013;347:f5001.Find this resource:
70. Brien SE, Ronksley PE, Turner BJ, Mukamal KJ, Ghali WA. Effect of alcohol consumption on biological markers associated with risk of coronary heart disease: systematic review and meta-analysis of interventional studies. BMJ 2011;342:d636.Find this resource:
71. Rimm EB, Klatsky A, Grobbee D, Stampfer MJ. Review of moderate alcohol consumption and reduced risk of coronary heart disease: is the effect due to beer, wine, or spirits? Br Med J 1996;312:731–736.Find this resource:
72. Friedman LA, Kimball AW. Coronary heart disease mortality and alcohol consumption in Framingham. Am J Epidemiol 1986;124:481–489.Find this resource:
73. Keil U, Chambless LE, Döring A, Filipiak B, Stieber J. The relation of alcohol intake to coronary heart disease and all-cause mortality in a beer-drinking population. Epidemiology 1997;8:150–156.Find this resource:
74. Gaziano JM, Gaziano TA, Glynn RJ, et al. Light-to-moderate alcohol consumption and mortality in the Physicians’ Health Study enrollment cohort. J Am Coll Cardiol 2000;35:96–105.Find this resource:
75. Thun MJ, Peto R, Lopez AD, et al. Alcohol consumption and mortality among middle-aged and elderly U.S. adults. N Eng J Med 1997;337:1705–1714.Find this resource:
76. Chen WY, Rosner B, Hankinson SE, Colditz GA, Willett WC. Moderate alcohol consumption during adult life, drinking patterns, and breast cancer risk. JAMA 2011;306:1884–1890.Find this resource:
77. Ding M, Satija A, Bhupathiraju SN, et al. Association of coffee consumption with total and cause-specific mortality in 3 large prospective cohorts. Circulation 2015;132:2305–2315.Find this resource:
78. Freedman ND, Park Y, Abnet CC, Hollenbeck AR, Sinha R. Association of coffee drinking with total and cause-specific mortality. N Engl J Med 2012;366:1891–1904.Find this resource:
79. Leung CW, Ding EL, Catalano PJ, Villamor E, Rimm EB, Willett WC. Dietary intake and dietary quality of low-income adults in the Supplemental Nutrition Assistance Program. Am J Clin Nutr 2012;96:977–988.Find this resource:
80. Morris MC, Tangney CC. A potential design flaw of randomized trials of vitamin supplements. JAMA 2011;305:1348–1349.Find this resource:
81. Grodstein F, Kang JH, Glynn RJ, Cook NR, Gaziano JM. A randomized trial of beta carotene supplementation and cognitive function in men: the Physicians’ Health Study II. Arch Intern Med 2007;167:2184–2190.Find this resource:
82. Gaziano JM, Sesso HD, Christen WG, et al. Multivitamins in the prevention of cancer in men: the Physicians’ Health Study II randomized controlled trial. JAMA 2012;308:1871–1880.Find this resource:
83. Trichopoulou A, Costacou T, Bamia C, Trichopoulos D. Adherence to a Mediterranean diet and survival in a Greek population. N Engl J Med 2003;348:2599–2608.Find this resource:
84. de Lorgeril M, Renaud S, Mamelle N, et al. Mediterranean alpha-linolenic acid-rich diet in secondary prevention of coronary heart disease [Erratum in: Lancet 1995;345:738]. Lancet 1994;343:1454–1459.Find this resource:
85. McCullough ML, Feskanich D, Stampfer MJ, et al. Diet quality and major chronic disease risk in men and women: moving toward improved dietary guidance. Am J Clin Nutr 2002;76:1261–1271.Find this resource:
86. Chiuve SE, Fung TT, Rimm EB, et al. Alternative dietary indices both strongly predict risk of chronic disease. J Nutr 2012;142:1009–1018.Find this resource:
87. Stampfer MJ, Hu FB, Manson JE, Rimm EB, Willett WC. Primary prevention of coronary heart disease in women through diet and lifestyle. N Engl J Med 2000;343:16–22.Find this resource:
88. Chomistek AK, Chiuve SE, Eliassen AH, Mukamal KJ, Willett WC, Rimm EB. Healthy lifestyle in the primordial prevention of cardiovascular disease among young women. J Am Coll Cardiol 2015;65:43–51.Find this resource:
89. Willett WC. Balancing life-style and genomics research for disease prevention. Science 2002;296:695–698.Find this resource:
90. Willett WC, Dietz WH, Colditz GA. Guidelines for healthy weight. N Engl J Med 1999;341:427–434.Find this resource:
91. Yuan C, Spiegelman D, Rimm EB, et al. Validity of a dietary questionnaire assessed by comparison with multiple weighed dietary records or 24-hour recalls. Am J Epidemiol; in press.Find this resource:
92. Rifas-Shiman SL, Willett WC, Lobb R, Kotch J, Dart C, Gillman MW. PrimeScreen, a brief dietary screening tool: reproducibility and comparability with both a longer food frequency questionnaire and biomarkers. Public Health Nutr 2001;4:249–254.Find this resource:
93. Willett WC.Eat, Drink, and Be Healthy. 2nd ed. New York, NY: Simon & Schuster; 2005.Find this resource:
94. Willett WC, Katzen M.Eat, Drink, and Weigh Less. New York, NY: Hyperion; 2006.Find this resource:
95. Appel LJ, Moore TJ, Obarzanek E, et al. A clinical trial of the effects of dietary patterns on blood pressure. DASH Collaborative Research Group. N Engl J Med 1997;336:1117–1124.Find this resource:
96. Group SR, Wright JT Jr, Williamson JD, et al. A randomized trial of intensive versus standard blood-pressure control. N Engl J Med 2015;373:2103–2116.Find this resource:
97. Sacks FM, Svetkey LP, Vollmer WM, et al. Effects on blood pressure of reduced dietary sodium and the dietary approaches to stop hypertension (DASH) diet. N Engl J Med 2001;344:3–10.Find this resource:
98. Ornish D, Scherwitz LW, Billings JH, et al. Intensive lifestyle changes for reversal of coronary heart disease. JAMA 1998;280:2001–2007.Find this resource:
99. Willett WC, Wood M, Childs D.Thinfluence. New York, NY: Rodale; 2014.Find this resource:
101. Thorndike AN, Riis J, Sonnenberg LM, Levy DE. Traffic-light labels and choice architecture: promoting healthy food choices. Am J Prev Med 2014;46:143–149.Find this resource: