​

Questions and answers about the WHI Dietary Modification (DM) Trial

These questions and answers supplement the February 8, 2006 JAMA publications about the effects of a low-fat eating pattern on breast cancer, colorectal cancer, and coronary heart disease.
  1. What was the purpose of the Dietary Modification (DM) Trial?
  2. What was the rationale for conducting the DM Trial?
  3. Who participated in the DM Trial?
  4. How was the DM Trial conducted?
  5. What was asked of both groups, the comparison and dietary change groups? How were the women followed?
  6. How long were DM Trial participants followed?
  7. What dietary changes were asked of the comparison group?
  8. What was asked of the dietary change (intervention) group?
  9. What dietary changes did DM Trial participants make?
  10. What characteristics were associated with better adherence in the dietary change group?
  11. Since the dietary change group did not achieve the goal for total fat intake, what does this say about the feasibility of low fat diets among the general public?
  12. What are the findings from the WHI DM Trial?
  13. Why is it important to follow women in the WHI Extension Study?
  14. Is it safe to follow a low fat dietary pattern? And what about women who are already eating low fat for health?
  15. What does the study mean for cancer prevention nutrition education programs such as five a day?
  16. How will these findings impact future guidelines and recommendations on fat intake and cholesterol, and fat intake and weight loss?
  17. The study only found "trends toward" lower LDL in women consuming less "bad" fats (saturated fats). How does this hold up against the therapeutic lifestyle changes (TLC) and other national heart disease prevention guidelines that indicate that eating less trans fatty acids and saturated fat will lower LDL-cholesterol?
1. What was the purpose of the Dietary Modification (DM) Trial?
The DM Trial researched the effect of a low-fat, high fruit, vegetable, and grain diet on breast cancer, colorectal cancer and heart disease in postmenopausal women. Intervention goals were to reduce fat intake to 20% of calories, increase fruits/vegetables to five or more servings daily and increase grain servings to six or more daily.
2. What was the rationale for conducting the DM Trial?
Observational data, supported by animal studies, suggested that a low-fat diet might reduce the risk of breast and colorectal cancers. Effects on heart disease were also suggested, particularly through reductions in saturated fat intake. In the WHI, reductions in saturated fat were presumed to go along with reduction in total fat, as had occurred in the Women’s Health Trial, a pilot study of the WHI DM Trial.
3. Who participated in the DM Trial?
48,835 postmenopausal women of multiple races and ethnicities and varying ages participated in the DM Trial.
4. How was the DM Trial conducted?
The DM Trial was a randomized controlled clinical trial, the most rigorous of research designs and often referred to as the gold standard.
Participants were randomly assigned to one of two treatment groups for follow-up:
  • Comparison (No Intervention) – 60% of participants
  • Dietary Change (Intervention) – 40% of participants
The uneven randomization (60% Comparison and 40% Dietary Change) provided a cost-effective design without compromising the ability of the DM Trial to address its research questions.
5. What was asked of both groups, the comparison and dietary change groups? How were the women followed?
All participants in the DM Trial were asked to do the following:
  • Provide health information updates every six months, including information about certain diseases (such as cancer and heart disease), hospitalizations and certain procedures. If these outcomes were reported, more detailed records were requested to verify the outcome reported by the study participant.
  • Provide mammogram reports from their health care provider every two years.
  • Provide dietary intake information at the start of the trial (baseline) and at year 1. After year 1, each woman’s dietary intake was assessed every three years. Dietary information was collected mostly by a food frequency questionnaire that had questions about a woman’s eating habits for the previous three months. Women who had blood collected and physical measures obtained at years 3, 6, and 9 also completed more detailed dietary records at these years.
  • Provide information about current medications and dietary supplements at baseline and years 1, 3, 6, and 9.
  • Come to the clinic for physical exams every year, including measurement of height, weight, waist and hip circumference, resting pulse and blood pressure. At the 3rd, 6th, and 9th years, physical measurements were collected from a subsample of the entire group. The subgroup included the same women each year (a cohort).
  • Provide blood samples at the start of the study (baseline) and the first year. At the 3rd, 6th, and 9th years, blood samples were collected from a subsample of the entire group. The subsample (a cohort) included the same women at each of these years.
  • The cohort at years 3, 6, and 9 included the same women who provided both blood samples and physical measurements.
  • Staff and investigators were uninformed about (blinded to) treatment assignment when follow-up information was being collected.
6. How long were DM Trial participants followed?
Participants in both the Comparison and Dietary Change groups were followed for 8.1 years on average.
7. What dietary changes were asked of the comparison group?
None. Women in the Comparison group were asked to maintain their usual eating habits. They received standard nutrition information when they were randomized into the DM Trial (USDA Dietary Guidelines for Americans 1990 and 1995).
 
Then why have a comparison group?
  • To compare the number of cases of breast cancer, colorectal cancer, heart disease against the number in the Dietary Change group (the intervention group).
  • To account for gradual diet changes that women may make for reasons not related to the DM intervention.
8. What was asked of the dietary change (intervention) group?
Women in the Dietary Change group were asked to decrease their fat intake to 20 percent of their total daily calories; increase fruits and vegetables combined to five or more servings per day; and increase grains to six or more servings per day.
Dietary Change participants were taught how to adhere to the recommended diet by changing their eating habits. They were assisted in doing this by:
  • Attending group sessions. Participants met in groups of 8-15 women 18 times during the first year and four times annually thereafter. These groups were led by registered dietitians or nutritionists. The women were provided information to assist them in making the dietary changes that were the goals of the trial.
  • Self-monitoring their intake of fat, fruits, vegetables, and grains. Self-monitoring is a well documented aid in changing behaviors and maintaining the changes. Women were asked to monitor their intake. The women were given information about fat grams and fruit/vegetable and grain servings in foods. They were also given food diaries or alternative self-monitoring tools for writing down what they ate.
9. What dietary changes did DM Trial participants make?
Total Fat

 
The Dietary Change (intervention) group achieved greater reductions in total fat intake and sustained them longer than in any previous dietary trial in history. The Dietary Change group went from 35% to 24% of calories from total fat the first year, to 29% the fifth, sixth, and seventh years. In the Comparison group the fat intake stayed at its baseline 35% calories from fat at after the first year and increased to 37% the fifth, sixth, and seventh years. Fat intake in the Dietary Change group was 11 percentage points lower in the first year and 8 percentage points lower in the fifth, sixth, and seventh years than in the Comparison group.

Women in the Dietary Change group achieved about 70% of the designed reduction in fat to a total of 20% calories from fat. However, relatively few women met the dietary target of 20% energy from fat: 31.4% of women at Year 1 and 14.4% of women at Year 6.
 
Sources of Dietary Fat

 
At the start of the DM Trial, the major sources of fat intake in both the Comparison and Dietary Change groups were added fats (butter, margarine, oils, salad dressings added during food preparation or at the table), meats, desserts, milk and cheese, mixed dishes, and high-fat breads and salty snacks.

After the first year of the study, sources of fat in the Comparison group did not change. In the Dietary Change group, women made the greatest changes by decreasing added fats. Fewer changes were made in other food sources of fat and meats became the major source of fat.
 
Fruits/Vegetables and Grains
The Dietary Change group increased fruit/vegetable intake per day from 3-1/2 to 5 servings by the first year, an increase that they maintained through the sixth year. The Comparison group increased their fruit/vegetable intake slightly from 3-1/2 servings at baseline to nearly 4 servings at the first and sixth years.
Grain intake, starting at slightly more than 4-1/2 servings daily at baseline, changed very little in either the Dietary Change or Comparison group.
10. What characteristics were associated with better adherence in the dietary change group?
Dietary change and maintaining change was strongly associated with attending the nutritionist-led group sessions and self-monitoring food intake. Women who reported better physical functioning, general health, and psychological well-being at the start of the trial reported lower fat intakes after the first year. Having additional contacts with a nutritionist also helped some women lower their fat intake.
11. Since the dietary change group did not achieve the goal for total fat intake, what does this say about the feasibility of low fat diets among the general public?
The results suggest that it is feasible to reduce fat intake to below 30% in large numbers of women, but that it will probably require more intensive or selective intervention to get it down to lower levels, e.g., 20%, and keep it there.
12. What are the findings from the WHI DM Trial?

Breast Cancer 

What were the effects of the low-fat dietary pattern on risk of breast cancer?
Overall, there was a non-significant 9% lower rate of breast cancer in women in the Dietary Change group compared to women in the Comparison group. This means that, out of 10,000 women, 42 women in the Dietary Change group and 45 in the Comparison group developed breast cancer each year.
If the study achieved its fat intake goal, would the result have been different? What effect did adherence to the low-fat dietary pattern have on the results?
The dietary change group achieved about 70% of the goal reduction in fat intake, and the 9% reduction in breast cancer is about 70% of that which had been predicted at the start of the study. There are reasons to think that the modest reduction observed in breast cancer occurrence in the Dietary Change group may not be due to chance alone since women who started at higher levels of fat intake (and who achieved more change) and women who were more adherent (completed the majority of trial activities) had the greatest reductions in breast cancer risk. The modest reduction in breast cancer risk, although not statistically significant after 8.1 years of follow-up, suggests a trend that with longer follow-up may yield a more definitive answer.
In light of other studies and the trends reported in the breast cancer paper, should women at risk of breast cancer eat a low fat diet?
The WHI results do not provide definitive evidence to guide women’s choices about the ideal diet to prevent breast cancer. There was a suggestion in the WHI trial that women who started with high fat intakes may be more successful in reducing their fat intake, and this may help reduce risk of breast cancer. Thus, women who are currently eating a diet high in fat may want to think seriously about decreasing their fat intake. However, it seems unlikely that changes in any one single food group or source of energy will improve overall health; rather, the total dietary pattern and lifestyle (including exercise) may be the key to better health.
Did the low fat dietary pattern affect the risk of breast cancer in women on hormone therapy?
The dietary pattern did not affect the risk of breast cancer in women who were on hormone therapy (either estrogen plus progestin or estrogen-alone) at baseline or from the Hormone Therapy Trials.
Did the effect of the dietary intervention on risk of breast cancer differ by tumor type, such as estrogen receptor status or progesterone receptor status?
Yes, women in intervention group experienced a significantly greater reduction in the risk of progesterone receptor negative breast tumors than in the risk of progesterone receptor positive tumors. It appeared that the low fat dietary pattern may have had the most benefit in preventing tumors that were both estrogen receptor positive and progesterone receptor negative. These findings support the argument of some effect of the intervention on breast cancer risk, and reinforce the view that breast cancer is a heterogeneous diagnosis with several major subtypes.
 
Colorectal Cancer

 
What were the effects of the low-fat dietary pattern on risk of colorectal cancer?

The WHI dietary intervention did not reduce the risk of colorectal cancer. The results were similar when looking at where the cancer occurred – the entire colon, upper or lower colon, or rectum.
What effect did adherence to the low-fat dietary pattern have on the results?
The results were similar when different levels of dietary adherence among Dietary Change participants were taken into consideration.
Is it possible that a benefit for colorectal cancer will emerge if a low fat dietary pattern is followed for a longer time?
Colorectal cancer can take decades to develop, and is thought to go through a stage of polyp or adenoma first. The reduced rates of self-reported polyp or adenoma in the Dietary Change group may result in delayed benefits in risk of colorectal cancer. A 5-year extended follow-up in these participants is ongoing and may elucidate this, even though the intervention phase of the DM Trial has ended and women are not being instructed to maintain their original dietary assignment (usual or low-fat dietary pattern).
What is the message for middle-age, older, or younger women?
Results from this paper suggest that changing dietary patterns, as the Dietary Change group did, in postmenopausal mid to later life may have limited benefit in preventing colorectal cancers in postmenopausal women.
The study does not answer questions about benefits or risks of changing dietary patterns among women of younger ages.
Was there any evidence of benefit among subgroups of women, for example women who exercise more?
It is too early to say; additional subgroup analyses will be performed.

Heart Disease
What were the effects of the low-fat dietary pattern on risk of heart disease?
The DM intervention did not reduce the risk of heart disease. However, the intervention was not designed specifically to study heart disease and did not focus on dietary components that are thought to influence coronary heart disease (CHD), such as reducing saturated and trans fat and increasing mono unsaturated and polyunsaturated fats. There were trends toward reduction in CHD risk in those who had the greatest decrease in saturated and trans fat.
Were their effects on risk factors associated with heart disease?
There were small but significant decreases in body weight, LDL-cholesterol and diastolic blood pressure. Triglycerides, HDL-cholesterol, glucose, and insulin were not increased by the diet intervention.
Overall
Taking the three outcomes together (breast cancer, colorectal cancer, and heart disease), is there any evidence of benefit for women who are able to change their eating pattern along the lines of the intervention?
Secondary analyses conducted in the three papers suggest that women who were consuming large amounts of fat and made larger reductions in fat intake experienced some benefit. These include a lower rate of breast cancer. They also suggested that women who lowered their total fat and reduced their saturated and trans fatty acids to a greater extent, or increased their intake of fruits and vegetables, experienced a lower rate of heart disease. Women using aspirin or combination hormones (estrogen plus progestin) might experience a reduction in colorectal cancer with a low fat dietary pattern. However, some of these analyses have not been completely explored may not have adequately controlled for confounding factors. More detailed analyses will be conducted in the future.
Were there any effects of the WHI low-fat dietary pattern on body weight?
The WHI low-fat dietary pattern was not designed for weight loss. However, many women in the Dietary Change group maintained or lost weight on the low-fat dietary pattern. Women in the dietary change group weighed about 5 pounds less than the comparison group at one year, and after 9 years of follow-up still weighed about 1 pound less. This overall slowing of weight gain is good news since obesity is on the rise in the United States.
Does this study have any implications for younger women?
The age range of women in this study was 50-79 years at baseline (start of the study; 1993-1998), and the results may not apply to younger women. It remains untested that a low fat dietary pattern started at a younger age and continued for many years has health benefits that could not be demonstrated over the 8 years of this study.
Were the findings for cancer and heart disease different in women who were taking hormone therapy or aspirin?
Women who were on hormone therapy or aspirin did not have a different result for breast cancer or heart disease. The data suggest a possible reduction in risk for colorectal cancer in Dietary Change women who were either taking aspirin or were on combination estrogen plus progestin hormone therapy; however, these findings are preliminary and need further review.
13. Why is it important to follow women in the WHI Extension Study?
Because the WHI intervention could have effects that do not appear until later, the continued follow-up of these women in the WHI Extension Study is expected to provide more conclusive evidence of the long term effects of a low-fat eating pattern initiated in middle age.
14. Is it safe to follow a low fat dietary pattern? And what about women who are already eating low fat for health?
Yes, a low-fat dietary pattern remains an option for generally healthy postmenopausal women. No adverse effects were found from the WHI low-fat dietary pattern. Further, the WH DM Trial results do not contradict the USDA Dietary Guidelines for Americans (www.healthierus.gov/dietaryguidelines), which advise a total fat intake of 20-35% of energy and full of fruits, vegetables, and whole grains. Thus, the Dietary Guidelines for Americans 2005 remain an option for generally healthy postmenopausal women. For heart disease, the Dietary Guidelines recommendations to reduce specific types of fat (saturated and trans fatty acids) and consume mostly polyunsaturated and monounsaturated fatty acids instead seem even more important based on the WHI DM Trial results.
Comparison of the WHI low-fat dietary pattern goals to the USDA Dietary Guidelines for Americans 2005
Low-fat Dietary Pattern from the WHI DM Trial USDA Dietary Guidelines for Americans 2005

Goals

Recommendations

Food examples
Total fat
20% of total calories		 Total fat
20-35% of total calories
Saturated fat: 7% or less of total calories** Saturated fat: 10% of total calories. Saturated fats: Dairy and animal fats, such as butter, most cheeses and full or low-fat fat dairy products
Other fats, such as polyunsaturated, monounsaturated, and trans: WHI did not have goals Trans fats: Keep as low as possible. Trans fats: Processed foods and oils containing hydrogenated oils, such as some shortenings and margarines used in food preparation or at the table.
Polyunsaturated and monounsaturated fats: Consume most fats from these sources. Polyunsaturated fats: Oils from fish, nuts, and vegetable oils such canola, corn, or soy oils.
Monounsaturated fats: Olives and olive oil, avocados, peanuts and peanut oils.
Fruits and/or vegetables
5 or more servings daily		 Fruits and/or vegetables
5-9 servings of fruits and vegetables, depending on calorie intake.		 Choose a variety of fruits and vegetables daily. Include a mix of dark green and orange fruits and vegetables, legumes, starchy vegetables and other vegetables several times a week.
Grains
6 or more servings daily**		 Grains
3 or more whole grain servings daily, depending on calorie intake.		 Whole grains typically consumed in the U.S. include whole wheat, oats or oatmeal, popcorn, brown rice, whole rye, whole grain barley, wild rice, buckwheat, triticale, bulgur, millet, quinoa, and sorghum.
**The WHI Dietary Study had a goal for saturated fat, but Dietary Change participants were not asked to record their intake of saturated fat intake because it usually drops as total fat is reduced. Whole grains were encouraged as part of a healthy eating pattern, but they were not a specific study goal.
The Dietary Guidelines are revised every five years based on the totality of evidence from many dietary studies. The Dietary Guidelines 2005 differ in several important respects from the WHI low-fat dietary pattern. The Guidelines emphasize reducing as much as possible the intakes of saturated fatty acids and trans fatty acids, with most fats in the diet coming from sources of polyunsaturated and monounsaturated fatty acids instead. These recommended sources can be found in foods including fish, nuts, and vegetable oils. In contrast, with the goal of reducing cancer risk, the WHI dietary pattern reduced all types of fats.
The Dietary Guidelines 2005 also include elements of the diet recommended in the successful Dietary Approaches to Treat Hypertension (DASH) Trial. The DASH diet was designed to lower blood pressure and included food choices aimed at decreasing sodium intake and increasing magnesium, calcium, and potassium intake. DASH emphasized decreasing salt intake to less than one teaspoon a day and increasing intake of fat-free and low-fat milk products, as well as vegetables and fruit. The WHI DM Trial aims differed from DASH, notably the WHI low-fat dietary pattern did not specifically target reductions in sodium intake or increases in low fat dairy products, and had only small effects on increasing vegetable and fruit intake. The DASH diet reduced blood pressure in both hypertensive and non-hypertensive individuals, while the WHI dietary pattern did not.
An individual’s personal health care providers, including registered dietitians, can offer guidance about a dietary pattern that is right for each individual. In addition, the American Dietetic Association (http://www.eatright.org), the nation’s largest organization of food and nutrition professionals, offers a toll-free consumer hotline (1-800-366-1655) of brief pre-recorded nutrition messages or help finding a registered dietitian in one’s local area.
15. What does the study mean for cancer prevention nutrition education programs such as five a day?
The WHI DM Trial was not designed to examine specifically the effects of vegetables and fruits, and the low-fat dietary pattern (the intervention) focused more on decreasing total fat intake. For cancer prevention, the WHI dietary pattern suggested a possible small benefit for breast cancer, but no benefit for colorectal cancer. As with eating low fat, eating more servings of vegetables and fruits, as recommended in the USDA Dietary Guidelines for Americans 2005 and by the national Five-A-Day program is not contraindicated by the WHI DM Trial results.
In general, fruits and vegetables can replace less healthy food choices such as refined carbohydrates and saturated fats. As shown by the DASH diet, an increase in vegetables and fruits, together with increase in low fat dairy products and a decrease in sodium, can help to reduce blood pressure. In WHI, there was a suggestion that women who chose to increase their vegetable and fruit intake to a greater extent had some reduction in risk of CHD.
16. How will these findings impact future guidelines and recommendations on fat intake and cholesterol, and fat intake and weight loss?
It seems unlikely that the current findings will change guidelines and recommendations. However, it may be too early to know as suggested by the importance of longer term follow-up, especially for breast and colorectal cancers. The low-fat dietary pattern used by WHI does not increase the risk of gaining weight, and there was no increase in triglycerides, blood glucose, or insulin—all of which suggest that such a diet will not increase the risk of heart disease or diabetes. The WHI DM Trial effect on diabetes has not yet been published. More understanding of the impact of the WHI DM Trial will come as further analyses are conducted.
17. The study only found "trends toward" lower LDL in women consuming less "bad" fats (saturated fats). How does this hold up against the therapeutic lifestyle changes (TLC) and other national heart disease prevention guidelines that indicate that eating less trans fatty acids and saturated fat will lower LDL-cholesterol?
The WHI low fat dietary pattern was aimed mainly at reducing the risk of cancer, and therefore emphasized total fat rather than type of fat. Many other trials have shown that changes in specific fats are needed to reduce LDL cholesterol—decreases in saturated fatty acids and trans fatty acids, with most of the fat in the diet coming from polyunsaturated and monounsaturated fatty acids. In WHI those women who made bigger changes in lowering saturated fatty acids and trans fatty acids had bigger changes in lowering LDL cholesterol, and in lowering risk of CHD. Therefore, the results of the WHI DM Trial do not contradict the current guidelines.

See also