By Ellen Feldman, MD

Altru Health System, Grand Forks, ND

Dr. Feldman reports no financial relationships relevant to this field of study.

SYNOPSIS: Investigators studied changes in quantity of nut consumption and relative risk of cardiovascular disease, finding significantly lower risk when nut consumption increases by > 0.5 servings daily.

SOURCE: Liu X, Guasch-Ferré M, Drouin-Chartier JP, et al. Changes in nut consumption and subsequent cardiovascular disease risk among U.S. men and women: 3 large prospective cohort studies. J Am Heart Assoc 2020:9.e013877.

“The doctor of the future will give no medicine, but will instruct his patient in the care of the human frame, in diet, and in the cause and prevention of disease.” — Thomas Edison, 1903

Welcome to “the future.” Although Thomas Edison’s remarks do not fully represent the state of medical practice today, our current emphasis on disease prevention and increased understanding of the central role of diet in such efforts reverberate with the second half of his prediction.1

According to nationally representative data collected between 1999 and 2015, diets in the United States are trending healthier.2 The Healthy Eating Index, an objective measure reflecting a nutritious diet, ticked upward from 55.7 to 57. 7 during this time, in large part because of decreased consumption of sugar-sweetened beverages and trans fatty acids, as well an increased intake of whole grains, nuts, and seeds.3

Previous studies have documented a link between the consumption of nuts and a decrease in both the risk of cardiovascular disease (CVD) and coronary heart disease (CHD) mortality.4 Liu et al attempted to broaden our knowledge of this relationship by studying changes in nut consumption over time and whether the specific type of nut is significant.

Respondents came from three large-scale U.S. prospective studies: 34,103 men from the Health Professionals Follow-up Study (HPFS),5 77,815 women from the Nurses Health Study (NHS), and 80,737 women from the Nurses Health Study 2 (NHS 2).6 These are long-term studies, with the HPFS beginning in 1986, the NHS in 1976, and the NHS 2 in 1989. As part of these studies, questionnaires were sent every two to four years to collect information about health conditions, medications, lifestyle, and diet. The food frequency questionnaires, sent every four years, began in 1986 for the HPFS and a little later for the NHS 1 and 2 — in 1991. Questions about specific nuts — walnuts, other tree nuts, and peanuts (despite peanuts being legumes, they are treated as nuts in the study) — were added in the late 1990s.5,6

For each four-year period, Liu et al analyzed changes in total nut consumption of specific types of nuts and then compared them with the risk of developing CVD, CHD, or stroke in the following four-year period. This continued over 26 years for participants in the HPFS and NHS and for 22 years for NHS 2. The follow-up periods ended between January 2012 and June 2013.

Defining one serving of nuts as one ounce (or 28 grams), five categories of nut-eating habits over each four-year period were identified. Categories ranged from “minimal change” from the previous four-year block to increase or decrease of 0.01-0.49 servings/day, respectively, and increase or decrease of 0.5 servings/day, respectively.

Excluded from the study were survey participants who reported CVD, cancer, diabetes, or who did not report body mass index during the baseline collection period (first four years.) Data were progressively adjusted for multiple factors, culminating in “model four,” which adjusted for variables including age, sex, race, family history of CVD, number of teeth, initial and any change in alcohol intake, initial and any change in the Healthy Eating Index score, and activity level.

Independent of initial nut consumption, an increase in nut consumption over four years of > 0.5 daily servings (where one serving = 28 grams) compared to individuals who did not change intake of nuts was associated with lowered risk of CVD (relative risk [RR], 0.86; 95% CI, 0.78-0.94; P trend < 0.0001), lowered risk of CHD (RR, 0.88; 95% CI, 0.78-1.00; P trend = 0.01), and lowered risk of stroke (RR, 0.82; 95% CI, 0.71-0.96; P trend = 0.0004).

Independent of initial nut consumption, a decrease in nut consumption over four years of < 0.5 daily servings (where one serving = 28 grams) compared to individuals who did not change their intake of nuts was associated with higher risk of CVD (RR, 1.14; 95% CI, 0.99-1.32; P trend < 0.0001), higher risk of CHD (RR, 1.06; P trend = 0.01), and higher risk of stroke (RR, 1.28; P trend = 0.0004).

Risk stratified for type of nut, increasing intake of tree nuts, walnuts, and peanuts, per 0.5 serving/day, was associated with a decreased risk of CVD (RR, 0.91; 95% CI, 0.88-0.96). Increasing intake of tree nuts and peanuts, per 0.5 serving/day, was associated with a decreased risk of CHD (RR, 0.90; 95% CI, 0.83-0.98). Increasing intake of walnuts, per 0.5 serving/day, was associated with a decreased risk of stroke (RR, 0.80; 95% CI, 0.67-0.95).

Risk for habitually high nut consumers > 0.5 servings/day when compared with nonconsumers: Lower risk for CVD (RR, 0.75; 95% CI, 0.67-0.84), lower risk for CHD (RR, 0.80; 95% CI, 0.69-0.93), and lower risk for stroke (RR, 0.68; 95% CI, 0.57-0.82).

In further analysis, Liu et al reported that additional lowering of CVD, CHD, and stroke risk is noted among a group of participants reporting simultaneous increase in nut consumption and decrease in intake of red meat.


Liu et al presented 26 years of follow-up data from three long-term, prospective studies. The results add strength to growing evidence that consuming nuts is healthy, and most likely cardioprotective. This study carries clear clinical relevance. The results suggest increasing nut consumption, even after years of less intake, still can lead to significant health benefits.

Although the association with decreased risk of CVD was stronger for consistent high-volume nut consumers, there still was a significant association with reduced risk in participants who reported recent such dietary changes. Also, when nuts replace less healthy dietary components (such as red meat), additional health benefits accrue. More studies should bring extra clarity to this aspect of dietary change. Unsaturated fatty acids, fiber, phytochemicals, antioxidants, and vitamins are among the components of nuts; all of these most likely contribute to a role in disease prevention. Other studies have shown decreased cholesterol and triglycerides associated with daily nut intake. Although nuts are relatively high in calories, previous work from Liu et al demonstrated a reduced risk of obesity over time associated with higher nut intake.7

Liu et al noted that previous studies of nut consumption and CVD and/or CHD have consistently shown an inverse association, but the same does not apply for studies of stroke risk and nut consumption. These latter studies have produced inconsistent results. Liu et al noted walnuts appear to be more strongly associated with lowering risk of stroke in this work than all tree nuts. They wondered if this is because, in part, of a suspected association of walnuts with blood-pressure lowering. This is an area “ripe” for future investigation.

One strength of this Liu et al study lies in the long follow-up period; another strength resides in the number of respondents. The dedication of the staff and participants in the HPFS and NHS 1 and 2 have resulted in invaluable contributions to our knowledge about health habits, prevention, and disease progression. However, one of the downsides of these efforts is respondent homogeneity. The members of the HPFS and NHS are health professionals, tend to be of European ancestry (although this is changing with the NHS 3), and tend to be healthier than the population at large.5,6 These factors make it difficult to confidently generalize the results from these studies.

Another relative weakness of this study was that all consumption data were self-reported, thus subject to recollection errors. Other researchers may want to track diet in a more sophisticated manner. Real-time information about diet can add specifics of how the nuts are prepared (raw, roasted, etc.). Additional information about serving size can assist in providing information about any plateau of health benefits in respect to quantity. All these areas could add to the goal of providing detailed clinical advice regarding this intervention.

Even with limitations, this study strengthens evidence for the health benefits of nuts. Notably, peanuts (a relatively affordable nut) were included among the choices and showed an association with health benefits. When working with patients on prevention, promoting the pivotal role of at least 0.5 servings daily of nuts (or 0.5 ounces) in a healthy diet and wellness plan is backed by solid medical evidence.


  1. U.S. Department of Health and Human Services. Food & nutrition.
  2. Shan Z, Rehm CD, Rogers G, et al. Trends in dietary carbohydrate, protein, and fat intake and diet quality among U.S. adults, 1999-2016. JAMA 2019;322:1178-1187.
  3. U.S. Department of Agriculture. Healthy Eating Index (HEI). Updated July 21, 2020.
  4. Martínez-González MA, Salas-Salvadó J, Estruch R, et al. Benefits of the Mediterranean Diet: Insights from the PREDIMED study. Prog Cardiovasc Dis 2015;58:50-60.
  5. Harvard T.H. Chan School of Public Health. Health professionals follow-up study.
  6. Nurses’ Health Study.
  7. Liu X, Li Y, Guasch-Ferré M, et al. Changes in nut consumption influence long-term weight change in U.S. men and women. BMJ Nutrition Prevention & Health 2019. doi: 10.1136/bmjnph-2019-000034