Abstract
Background
The associations between specific types of nuts, specifically peanuts and walnuts, and cardiovascular disease remains unclear.
Objectives
To analyze the associations between the intake of total and specific types of nuts and cardiovascular disease, coronary heart disease, and stroke risk.
Methods
We included 76,364 women from the Nurses’ Health Study (1980–2012), 92,946 women from the Nurses’ Health Study II (1991–2013), and 41,526 men from the Health Professionals Follow-up Study (1986–2012) who were free of cancer, heart disease, and stroke at baseline. Nut consumption was assessed using food frequency questionnaires at baseline and updated every 4 years.
Results
During 5,063,439 person-years of follow-up, we documented 14,136 incident cardiovascular disease cases, including 8,390 coronary heart disease cases and 5,910 stroke cases. Total nut consumption was inversely associated with total cardiovascular disease and coronary heart disease after adjustment for cardiovascular risk factors. The pooled multivariable hazard ratios for cardiovascular disease and coronary heart disease among participants who consumed one serving of nuts (28 g) five or more times per week, compared to the reference category (never or almost never), were 0.86 [95% confidence interval (CI), 0.79–0.93, P trend 0.0002] and 0.80 [95% CI, 0.72–0.89, P trend <0.001], respectively. Consumption of peanuts and tree nuts (two or more times/week) and walnuts (one or more times/week), was associated with a 13%-19% lower risk of total cardiovascular disease and 15–23% lower risk of coronary heart disease.
Conclusions
In 3 large prospective cohort studies, higher consumption of total and specific types of nuts was inversely associated with total cardiovascular disease and coronary heart disease.
Keywords: Nuts, cardiovascular disease, coronary heart disease, stroke, peanuts, tree nuts
Introduction
Cardiovascular disease (CVD) remains a leading cause of death worldwide and hence prevention of CVD has become a top public health priority (1). In recent years, dietary recommendations have shifted towards diets high in plant-based foods and low in animal-based foods for the prevention of chronic diseases (2). Most of these plant-based dietary patterns highlight the intake of nuts as a key component. Nuts have a unique nutritional composition (3) and are good sources of unsaturated fatty acids, dietary fiber, minerals, vitamins, and other bioactive compounds (4).
Frequent nut consumption has been associated with reduced cardiovascular risk factors including dyslipidaemia, type 2 diabetes and metabolic syndrome; as well as with lower risk of coronary heart disease (CHD) (5–8). Findings from prospective cohort studies were confirmed by a randomized primary-prevention trial conducted in a Mediterranean population at high cardiovascular risk (9). Participants randomized to a Mediterranean diet supplemented with mixed nuts—hazelnuts, almonds, and walnuts—had a 28% reduction in the incidence of major cardiovascular events after about 5 years of follow-up (9). Recent findings from the Nurses’ Health Study (NHS) and the Health Professionals Follow-up Study (HPFS) have also provided further evidence that the frequency of nut consumption is inversely associated with total and cause-specific mortality (10).
Most of the previous prospective studies have focused on total nut consumption in relation to the risk of CVD (11, 12). However, the associations between peanut butter and specific types of nuts, such as peanuts and walnuts, with major cardiovascular events, and specifically the relation with stroke remain unclear. Of note, because the nutritional composition of peanuts and walnuts differs from other nuts (13), evaluating the health effects of specific types of nuts is of particular interest.
Our primary hypothesis is that frequent nut consumption is associated with lower risk of CVD incidence. Therefore, we examined the associations between the intake of total and specific types of nuts with CVD in three large prospective cohort studies. Our cohorts provided repeated measures of diet (including separate data on peanuts, walnuts, tree nuts, and peanut butter), extensive data on cardiovascular risk factors, up to 32 years of follow-up, and a large number of incident CVD cases.
Methods
Study population
The NHS is a prospective cohort study of 121,700 female nurses, aged 30 to 55 years, from 11 U.S. states that began in 1976. The NHS II was established in 1989 and consists of 116,671 younger female registered nurses, aged 25 to 42 years at baseline. The HPFS is a prospective cohort study of 51,529 male health professionals aged 40 to 75 years at baseline that began in 1986. In all three cohorts, information about medical history, lifestyle, and health conditions has been collected by self-administered questionnaires every two years since baseline. Detailed information on the cohorts has been described previously elsewhere (14–16).
For the present analysis, baseline year was defined as the first year when the validated food-frequency questionnaire (FFQ) was collected in each study—1980 for the NHS, 1991 for NHS II, and 1986 for HPFS. We excluded from the analysis those participants who reported CVD or cancer at baseline, participants who did not provide information on nut consumption, those who left more than 70 food items blank in the FFQ, or had daily energy intakes <600 or >3500 kcal for women and <800 or >4200 kcal for men. The final analyses included 76,364 women in the NHS, 92,946 women in the NHS II and 41,526 men in the HPFS. The institutional review boards of Brigham and Women’s Hospital and Harvard T.H. Chan School of Public Health approved the study protocol.
Ascertainment of CVD
Our primary outcome measure was major CVD defined as a combined endpoint of myocardial infarction, stroke, or fatal CVD (fatal stroke, fatal myocardial infarction, and CV death). We further assessed the following secondary outcome measures: total CHD, which was defined as fatal or non-fatal myocardial infarction; and total stroke, which included all fatal and non-fatal stroke cases (ischemic, hemorrhagic, and undetermined subtypes). When a participant (or family members of deceased participants) reported an incident event, permission was requested to examine their medical records by physicians who were blinded to participant risk factor status. For each endpoint, the month and year of diagnosis were recorded as the diagnosis date. Nonfatal events were confirmed through review of medical record by study investigators blinded to participant risk factor status. Myocardial infarction was confirmed if the World Health Organization criteria were met based on symptoms plus diagnostic electrocardiogram changes or elevated cardiac enzymes (17). If medical records were unavailable, we considered myocardial infarctions probable when the participant provided additional confirmatory information. Information on angina and coronary revascularization procedures (percutaneous transluminal coronary angioplasty or coronary artery bypass grafting surgery) was self-reported, and we included only events that occurred before a manifest cardiovascular event.
Strokes were confirmed if data in the medical records fulfilled the National Survey of Stroke criteria requiring evidence of a neurologic deficit with sudden or rapid onset that persisted for >24 hours or until death (18). We excluded cerebrovascular pathology due to infection, trauma, or malignancy, as well as “silent” strokes discovered only by radiologic imaging. Radiology reports of brain imaging (computed tomography or magnetic resonance imaging) were available in 89% of those with medical records. We classified strokes as ischemic stroke (thrombotic or embolic occlusion of a cerebral artery), hemorrhagic stroke (subarachnoid and intraparenchymal hemorrhage), or stroke of probable/unknown subtype (a stroke was documented but the subtype could not be ascertained owing to medical records being unobtainable).
Deaths were identified by reports of families, the U.S. postal system, or using death certificates obtained from state vital statistics departments and the National Death Index and confirmed through review of medical records or autopsy reports. Follow-up for deaths was >98% complete (19). Fatal CVD was defined as fatal CHD disease, fatal stroke, or fatal CVD. Fatal CHD was defined as ICD-9 (international classification of diseases, ninth revision) codes 410–412 and was considered confirmed if fatal CHD was confirmed via medical records or autopsy reports or if CHD was listed as the cause of death on the death certificate and there was prior evidence of CHD in the medical records. We designated as probable those cases in which CHD was the underlying cause on the death certificates but no prior knowledge of CHD was indicated and medical records concerning the death were unavailable. Similarly, we used ICD-9 codes 430–434 to define fatal stroke and followed the same procedures to classify cases of confirmed or probable fatal stroke. Lastly, fatal CVD was defined by ICD-9 codes 390–458.
Dietary assessment
A semi-quantitative food-frequency questionnaire with over 130 items administered every 2 to 4 years was used to assess dietary intake. In the 1980 and 1984 dietary questionnaires, we asked participants how often they had consumed a serving of nuts (serving size, 28 g [1 oz]) during the preceding year: never or almost never, one to three times a month, once a week, two to four times a week, five or six times a week, once a day, two or three times a day, four to six times a day, or more than six times a day. In subsequent food-frequency questionnaires, the question regarding nuts was split into two items: peanuts and other nuts. A specific question about walnut consumption was first introduced on the questionnaires in 1998 in the NHS and HPFS and 1999 in the NHS II. Total nut consumption was defined as the intake of peanuts, other nuts, and walnuts (if available). Intake of peanut and peanut butter or other forms of peanuts was assessed and analyzed separately in the current analysis. Tree nuts included all types of tree nuts including walnuts (not including peanuts, which are botanically a legume but with a similar fatty acid and nutrient profile as other nuts). Consumption of peanut butter was assessed in 1980, 1984, 1986, 1990, and 1994, with the same 9 responses as those for nut consumption (serving size, 15 g [1 tablespoon]). A validation study indicated that nut consumption was reported with reasonable accuracy; the corrected coefficient was 0.75 between the FFQ and four 1-week diet record for total nuts and peanut butter (20).
Statistical analysis
We calculated each individual’s person-time from the date of the return of the baseline questionnaire to the date of death or the end of follow-up (June 2012 in NHS, June 2013 in NHSII and January 2012 in HPFS), whichever comes first. The cumulative average of nut consumption was calculated to better represent long-term diet and to minimize within-person variation. We stopped updating dietary variables on a report of cancer, coronary artery bypass, or angina because changes in diet after development of these conditions may confound the relation between diet and chronic diseases.
We used Cox proportional-hazards regression models to estimate hazard ratios and 95% confidence intervals of developing total CVD, CHD, and stroke according to nut consumption categories. Separate analyses were conducted for fatal and nonfatal CVD, nonfatal myocardial infarction, fatal CHD, fatal and nonfatal stroke, and ischemic stroke. We did not analyze hemorrhagic stroke separately due to the lower number of cases. Multivariable models were adjusted for updated over time covariates: age (continuous); Caucasian (yes/no); body-mass index (<23, 23–24.9, 25–29.9, 30–34.9, ≥35 kg/m2); physical activity (metabolic-equivalents/week, quintiles); smoking status (never, past, current 1–14 cigarettes/d, current 15–24 cigarettes/d, current ≥25 cigarettes/d); physical examination for screening purposes (yes/no); current multivitamin use(yes/no); and current aspirin use (yes/no); family history of diabetes mellitus (yes/no), myocardial infarction (yes/no), or cancer (yes/no); history of diabetes mellitus (yes/no), hypertension (yes/no), or hypercholesterolemia (yes/no); intakes of total energy (kcal/d), alcohol (g/d), red or processed meat (servings/d), fruits (servings/d), and vegetables (servings/d) (all in quintiles); and, in women, menopausal status and hormone use (premenopausal, postmenopausal never users, postmenopausal past users, postmenopausal current users). In NHS II, the multivariable model was further adjusted for oral contraceptive use (never, past and current users). The above covariates were updated every 2 or 4 y using the most recent data for each 2-y follow-up interval. P values for trend were calculated with the use of the Wald test of a continuous variable based on the median number of servings of nuts consumed per day for each category.
We performed separate analyses for peanuts, tree nuts, walnuts, and peanut butter intake. Peanut butter models were additionally adjusted for quintiles of glycemic load, white bread, and soda intake, to control for dietary components associated with the consumption of peanut butter. For these analyses, we combined categories of high nut intake (≥2 servings/week) to maintain statistical power. We also analyzed nuts and types of nuts as a continuous variable (per 1 serving (28g) increase).
We conducted several sensitivity analyses to test the robustness of the results. First, we conducted pre-specified subgroup analyses by potential effect modifiers and the interaction between nut intake and covariates was examined using the likelihood-ratio test. Second, to test if our results were biased by selectively stopping updating diet after an intermediate outcome, we continuously updated diet until the end of follow-up. Third, instead of using repeated measures of diet, we used the most recent measure of diet. Fourth, we analyzed the association between baseline nut consumption and the incidence of CVD, CHD and stroke. Fifth, we further adjusted the models for the number of natural teeth (0–10, 11–16, 17–32 teeth) in NHS and HPFS, where the information was available, as tooth loss may be a marker for periodontal disease, which has been shown to be previously associated with CVD risk (21). Sixth, we further adjusted the models for the Alternate Healthy Eating Index (AHEI) (22), excluding the nut component. For the analysis including specific types of nuts, we mutually adjusted for other types of nuts (i.e., peanuts mutually adjusted for tree nuts and peanut butter). Finally, we conducted sensitivity analysis excluding BMI from the models.
The hazard ratios from multivariable models in each cohort were pooled with the use of an inverse variance–weighted meta-analysis using a fixed-effects model. P values for heterogeneity were calculated with the use of the Q statistics. Analyses were performed with the SAS statistical package (version 9.4, SAS institute). Statistical tests were two sided, and P values of less than 0.05 were considered to indicate statistical significance.
Results
During an average of 28.7 years of follow-up in the NHS, 21.5 years in the NHS II, and 22.5 years in the HPFS (and a total of 5,063,439 person-years across the three cohorts), we documented 14,136 incident CVD cases, including 8,390 CHD and 5,910 stroke cases. Compared to those participants who never or almost never consumed nuts, those who consumed nuts more frequently were older, had a lower BMI, were less likely to smoke, more likely to exercise, and consumed more fruits and vegetables (Table 1). Participants’ characteristics according to the frequency of peanut and walnut consumption are shown in Online Tables 1 and 2, respectively. Online Table 3 shows the Pearson correlations among different types of nuts in the three cohorts.
Table 1.
Characteristics of Person-Years according to Frequency of Nut Consumption
Frequency of Nut Consumption | |||||
---|---|---|---|---|---|
| |||||
Characteristics | Never or almost never | Less than once per week | Once per week | Two to four times per week | Five or more times per week |
Nut intake (servings/d) | 0 | 0.01–0.09 | 0.10–0.19 | 0.20–0.59 | ≥0.60 |
| |||||
Nurses’ Health Study (1980) | |||||
| |||||
Age (y) | 56.2(11.1) | 59.5(10.9) | 60.5(11.0) | 62.1(11.1) | 65.2(11.8) |
BMI (kg/m2) | 26.1(5.4) | 26.1(5.2) | 25.8(5.0) | 25.4(4.8) | 25.0(4.8) |
Physical activity (metabolic equivalents/wk) | 14.7(22.7) | 16.3(21.5) | 18.1(22.4) | 19.7(24.0) | 21.3(24.3) |
Family history of diabetes mellitus (%) | 28.3 | 28.8 | 28.2 | 28.0 | 26.4 |
Family history of myocardial infarction (%) | 19.5 | 19.0 | 18.5 | 18.1 | 17.8 |
Current smoker (%) | 19.5 | 15.0 | 13.3 | 13.1 | 12.7 |
Current menopausal hormone use (%) | 15.9 | 21.8 | 22.9 | 22.8 | 18.2 |
Baseline hypertension (%) | 40.2 | 39.1 | 37.3 | 36.0 | 34.6 |
Baseline hypercholesterolemia (%) | 36.8 | 42.0 | 42.4 | 41.7 | 39.6 |
Baseline diabetes (%) | 7.9 | 6.7 | 6.3 | 6.0 | 6.4 |
Multivitamin supplement use (%) | 43.5 | 50.5 | 54.1 | 57.6 | 60.5 |
Current aspirin use (%) | 43.5 | 48.2 | 49.2 | 50.7 | 48.0 |
Total energy intake (kcal/day) | 1524(499) | 1643(507) | 1780(530) | 1873(552) | 1956(582) |
Alcohol intake (g/d) | 4.0(8.6) | 4.7(9.1) | 5.6(9.6) | 6.1(10.2) | 6.6(11.0) |
Red or processed meat intake (servings/d) | 1.3(0.8) | 1.2(0.7) | 1.3(0.7) | 1.2(0.7) | 1.2(0.8) |
Fruit intake (servings/d) | 2.0(1.3) | 2.1(1.2) | 2.2(1.2) | 2.4(1.2) | 2.5(1.4) |
Vegetable intake (servings/d) | 2.1(1.1) | 2.3(1.1) | 2.5(1.1) | 2.7(1.2) | 2.7(1.3) |
Peanuts intake (servings/d) | 0 (0.1) | 0 (0.1) | 0.1(0.1) | 0.2(0.2) | 0.3(0.4) |
Tree nuts intake (servings/d) | 0 (0) | 0 (0.1) | 0.1(0.1) | 0.1(0.2) | 0.2(0.3) |
Walnuts intake (servings/d) | 0 (0) | 0.0(0.1) | 0.0(0.1) | 0.1(0.1) | 0.1(0.2) |
Peanut butter intake (servings/d) | 0.2(0.3) | 0.2(0.3) | 0.2(0.3) | 0.3(0.3) | 0.3(0.4) |
AHEI score (without nuts) | 46.4(9.8) | 47.1(9.3) | 48.1(9.3) | 49.3(9.5) | 50.9(9.6) |
| |||||
Nurses’ Health Study II (1991) | |||||
| |||||
Age (y) | 43.2(7.5) | 47.5(7.4) | 47.0(7.9) | 49.5(7.6) | 50.2(7.6) |
BMI (kg/m2) | 26.8(6.4) | 26.8(6.3) | 26.4(6.2) | 25.8(6.0) | 25.2(5.7) |
Physical activity (metabolic equivalents/wk) | 19.5(25.4) | 20.1(26.4) | 21.7(27.3) | 24.4(31.8) | 26.7(35.5) |
Family history of diabetes mellitus (%) | 34.2 | 34.5 | 34.1 | 34.2 | 34.9 |
Family history of myocardial infarction (%) | 37.6 | 37.1 | 36.8 | 36.5 | 34.5 |
Current smoker (%) | 9.0 | 8.5 | 9.1 | 8.6 | 9.6 |
Current menopausal hormone use (%) | 8.8 | 9.2 | 9.2 | 9.2 | 9.6 |
Baseline hypertension (%) | 19.3 | 19.9 | 18.8 | 17.5 | 15.4 |
Baseline hypercholesterolemia (%) | 30.8 | 32.3 | 31.6 | 31.5 | 29.7 |
Baseline diabetes (%) | 1.0 | 0.9 | 0.9 | 0.8 | 1.0 |
Multivitamin supplement use (%) | 33.1 | 35.2 | 34.8 | 36.5 | 38.5 |
Current aspirin use (%) | 24.0 | 29.2 | 30.0 | 30.8 | 30.1 |
Total energy intake (kcal/day) | 1646(527) | 1803(547) | 1965(564) | 2088(578) | 2266(602) |
Alcohol intake (g/d) | 3.3(6.8) | 4.3(7.9) | 5.2(8.8) | 5.7(9.4) | 5.6(9.9) |
Red or processed meat intake (servings/d) | 0.9(0.6) | 0.9(0.6) | 1.0(0.6) | 1.0(0.6) | 0.9(0.8) |
Fruit intake (servings/d) | 1.1(0.9) | 1.2(0.8) | 1.3(0.9) | 1.5(0.9) | 1.7(1.2) |
Vegetable intake (servings/d) | 3.1(1.9) | 3.5(1.8) | 3.8(2.0) | 4.3(2.2) | 4.7(2.6) |
Peanuts intake (servings/d) | 0 (0) | 0 (0) | 0.1(0.0) | 0.2(0.1) | 0.4(0.4) |
Tree nuts intake (servings/d) | 0 (0) | 0 (0) | 0.1(0.1) | 0.2(0.1) | 0.6(0.4) |
Walnuts intake (servings/d) | 0 (0) | 0 (0) | 0(0.1) | 0.1(0.1) | 0.2(0.5) |
Peanut butter intake (servings/d) | 0.1(0.2) | 0.2(0.2) | 0.2(0.2) | 0.2(0.3) | 0.3(0.4) |
AHEI score (excluding nuts) | 46.4(9.7) | 46.7(9.3) | 47.5(9.5) | 49.5(9.6) | 51.6(9.9) |
| |||||
Health Professional’s Follow-up Study (1986) | |||||
| |||||
Age (y) | 60.3(11.2) | 62.6(10.7) | 61.8(10.8) | 63.2(10.9) | 64.4(10.9) |
BMI (kg/m2) | 25.7(3.5) | 26.0(3.5) | 26.0(3.5) | 25.9(3.6) | 25.5(3.4) |
Physical activity (metabolic equivalents/wk) | 28.0(36.7) | 33.2(39.2) | 35.3(41.1) | 37.6(41.9) | 39.0(44.3) |
Family history of diabetes mellitus (%) | 17.6 | 20.0 | 18.9 | 19.4 | 19.0 |
Family history of myocardial infarction (%) | 32.3 | 30.8 | 30.8 | 30.5 | 30.9 |
Current smoker (%) | 8.1 | 5.7 | 6.4 | 6.2 | 7.0 |
Baseline hypertension (%) | 34.8 | 35.8 | 35.2 | 35.1 | 32.1 |
Baseline hypercholesterolemia (%) | 33.7 | 41.6 | 40.7 | 40.9 | 37.1 |
Baseline diabetes (%) | 5.9 | 6.0 | 5.8 | 6.2 | 6.3 |
Multivitamin supplement use (%) | 38.6 | 46.5 | 46.2 | 49.3 | 50.3 |
Current aspirin use (%) | 45.1 | 56.5 | 55.2 | 56.7 | 54.2 |
Total energy intake (kcal/day) | 1768(564) | 1837(567) | 1980(593) | 2136(624) | 2363(668) |
Alcohol intake (g/d) | 9.5(14.2) | 10.4(14.3) | 11.6(14.9) | 12.8(15.8) | 14.1(17.3) |
Red or processed meat intake (servings/d) | 1.0(0.8) | 1.0(0.7) | 1.1(0.7) | 1.1(0.8) | 1.1(0.8) |
Fruit intake (servings/d) | 2.2(1.6) | 2.2(1.3) | 2.4(1.4) | 2.5(1.5) | 2.8(1.7) |
Vegetable intake (servings/d) | 2.8(1.6) | 2.9(1.5) | 3.1(1.5) | 3.3(1.6) | 3.6(1.8) |
Peanuts intake (servings/d) | 0 (0) | 0 (0) | 0.1(0.0) | 0.2(0.1) | 0.7(0.5) |
Tree nuts intake (servings/d) | 0 (0) | 0 (0) | 0.1(0.0) | 0.1(0.1) | 0.4(0.4) |
Walnuts intake (servings/d) | 0.0(0.1) | 0.0(0.1) | 0.0(0.1) | 0.1(0.1) | 0.1(0.2) |
Peanut butter intake (servings/d) | 0.2(0.4) | 0.2(0.3) | 0.2(0.3) | 0.3(0.4) | 0.3(0.5) |
AHEI score (excluding nuts) | 42.5(10.1) | 43.4(9.2) | 44.1(9.1) | 45.3(9.0) | 47.4(9.4) |
Values are means (SD) or percentages. All variables except age are age-standardized. Frequency of nut consumption pertains to one serving of nuts, defined as 28 g. AHEI, Alternate Healthy Eating Index.
Age-adjusted and multivariable-adjusted analyses showed a consistent significant inverse association between total nut consumption and total CVD and CHD (Table 2). The pooled multivariable hazard ratios, without heterogeneity by sex or cohort, for those who consumed nuts five or more times per week (1 serving of nuts = 28g), as compared with those who never consumed nuts, were 0.86 (95% confidence interval [CI], 0.79–0.93, P<0.001) for total CVD and 0.80 (95% CI, 0.72–0.89, P<0.001) for CHD. Each serving increase of nuts was associated with 6% [HR: 0.94; 95% CI, 0.89–0.99)] and 13% [HR: 0.87; 95% CI, 0.81–0.94)] lower risk of CVD and CHD, respectively. Total nut consumption was inversely associated with both fatal [HR:0.76; 95% CI, 0.70–0.84, P=0.0003] and nonfatal CVD [HR:0.91; 95% CI, 0.85–0.98, P=0.005] in the pooled analysis of the three cohorts (Online Table 4).
Table 2.
Relative risk (95% CI) of cardiovascular events according to Frequency of Nut Consumption
Frequency of Nut Consumption | ||||||||
---|---|---|---|---|---|---|---|---|
| ||||||||
Never or almost never | Less than once per week | Once per week | Two to four times per week | Five or more Times per week | P value for trend | HR (95% CI) per 28 g increase | ||
TOTAL CARDIOVASCULAR DISEASE: fatal and nonfatal myocardial infarction + fatal and nonfatal stroke | ||||||||
| ||||||||
NHS | Median (g/d) | 0 | 1.68 | 3.92 | 9.24 | 27.0 | ||
Person-years | 384,33 | 972,973 | 391,790 | 331,935 | 117,245 | 2,198,274 | ||
Cases | 1,278 | 3,005 | 1,131 | 972 | 341 | 6,727 | ||
Crude Incidence/100000PY | 333 | 309 | 289 | 293 | 291 | |||
Age-adjusted model | 1 | 0.76 (0.71–0.81) | 0.69 (0.63–0.74) | 0.64 (0.58–0.69) | 0.62 (0.54–0.71) | <0.001 | 0.68 (0.59–0.77) | |
Multivariable-adjusted | 1 | 0.88 (0.82–0.94) | 0.85 (0.78–0.93) | 0.82 (0.75–0.90) | 0.85 (0.75–0.97) | 0.01 | 0.96 (0.90–1.00) | |
| ||||||||
NHS II | Median (g/d) | 0 | 1.47 | 3.92 | 7.98 | 23.03 | ||
Person-years | 706,91 | 733,092 | 325,741 | 204,999 | 29,959 | 2,000,710 | ||
Cases | 624 | 740 | 322 | 204 | 25 | 1,915 | ||
Crude Incidence/100000PY | 88 | 101 | 99 | 100 | 83 | |||
Age-adjusted model | 1 | 0.92 (0.82–1.03) | 0.92 (0.80–1.05) | 0.82 (0.70–0.97) | 0.66 (0.44–0.98) | <0.001 | 0.60 (0.43–0.83) | |
Multivariable -adjusted | 1 | 0.91 (0.82–1.02) | 0.95 (0.83–1.10) | 0.89 (0.75–1.06) | 0.75 (0.50–1.13) | 0.15 | 0.69 (0.49–0.98) | |
| ||||||||
HPFS | Median (g/d) | 0 | 1.96 | 3.92 | 7.84 | 24.08 | ||
Person-years | 121,40 | 218,753 | 209,348 | 230,225 | 84,725 | 864,455 | ||
Cases | 885 | 1,406 | 1,251 | 1,390 | 562 | 5,494 | ||
Crude Incidence/100000PY | 729 | 646 | 598 | 604 | 663 | |||
Age-adjusted model | 1 | 0.88 (0.80–0.95) | 0.86 (0.78–0.93) | 0.80 (0.73–0.87) | 0.78 (0.70–0.87) | <0.001 | 0.87 (0.80–0.94) | |
Multivariable -adjusted | 1 | 0.94 (0.86–1.02) | 0.93 (0.85–1.01) | 0.89 (0.81–0.97) | 0.87 (0.78–0.97) | 0.02 | 0.93 (0.85–1.00) | |
| ||||||||
Pooled* | Multivariable-adjusted | 1 | 0.91 (0.86–0.95) | 0.90 (0.85–0.95) | 0.86 (0.81–0.91) | 0.86 (0.79–0.93) | 0.0002 | 0.94 (0.89–0.99) |
| ||||||||
CORONARY HEART DISEASE: fatal and nonfatal myocardial infarction | ||||||||
| ||||||||
NHS | Person-years | 384,646 | 973,970 | 392,199 | 332,232 | 117,369 | 2,200,416 | |
Cases | 783 | 1,609 | 560 | 460 | 140 | 3,552 | ||
Crude Incidence/100000PY | 204 | 165 | 143 | 139 | 119 | |||
Age-adjusted model | 1 (Ref.) | 0.70 (0.65–0.77) | 0.60 (0.54–0.67) | 0.54 (0.48–0.61) | 0.48 (0.40–0.58) | <0.001 | 0.68 (0.59–0.77) | |
Multivariable -adjusted | 1 (Ref.) | 0.84 (0.76–0.91) | 0.76 (0.68–0.86) | 0.73 (0.64–0.82) | 0.69 (0.56–0.83) | <0.001 | 0.84 (0.74–0.94) | |
| ||||||||
NHS II | Person-years | 707,318 | 733,533 | 325,956 | 205,132 | 29,972 | 2,001,911 | |
Cases | 212 | 282 | 98 | 71 | 7 | 670 | ||
Crude Incidence/100000PY | 30 | 38 | 30 | 35 | 23 | |||
Age-adjusted model | 1 (Ref.) | 0.93 (0.77–1.12) | 0.74 (0.58–0.94) | 0.72 (0.55–0.95) | 0.46 (0.21–0.97) | 0.001 | 0.37 (0.20–0.69) | |
Multivariable -adjusted | 1 (Ref.) | 0.93 (0.77–1.13) | 0.79 (0.61–1.02) | 0.84 (0.63–1.12) | 0.57 (0.27–1.23) | 0.06 | 0.51 (0.27–0.95) | |
| ||||||||
HPFS | Person-years | 121,568 | 219,074 | 209,641 | 230,514 | 84,833 | 865,630 | |
Cases | 693 | 1,071 | 934 | 1,046 | 424 | 4,168 | ||
Crude Incidence/100000PY | 571 | 490 | 446 | 454 | 500 | |||
Age-adjusted model | 1 (Ref.) | 0.86 (0.78–0.94) | 0.82 (0.74–0.90) | 0.77 (0.70–0.85) | 0.76 (0.67–0.86) | <0.001 | 0.85 (0.77–0.93) | |
Multivariable -adjusted | 1 (Ref.) | 0.93 (0.84–1.03) | 0.90 (0.81–0.99) | 0.88 (0.79–0.97) | 0.86 (0.76–0.98) | 0.03 | 0.91 (0.82–0.99) | |
| ||||||||
Pooled* | Multivariable-adjusted | 1 (Ref.) | 0.88 (0.83–0.94) | 0.83 (0.78–0.90) | 0.82 (0.76–0.88) | 0.80 (0.72–0.89) | <0.0001 | 0.87 (0.81–0.94) |
| ||||||||
STROKE: fatal and nonfatal stroke | ||||||||
| ||||||||
NHS | Person-years | 384,731 | 973,952 | 392,151 | 332,238 | 117,328 | 2,200,400 | |
Cases | 536 | 1,462 | 590 | 529 | 205 | 3,322 | ||
Crude Incidence/100000PY | 139 | 150 | 151 | 159 | 175 | |||
Age-adjusted model | 1 (Ref.) | 0.87 (0.78–0.96) | 0.83 (0.74–0.93) | 0.79 (0.70–0.89) | 0.82 (0.69–0.89) | 0.04 | 0.96 (0.89–1.04) | |
Multivariable-adjusted | 1 (Ref.) | 0.97 (0.87–1.07) | 0.98 (0.86–1.10) | 0.96 (0.85–1.09) | 1.05 (0.88–1.26) | 0.50 | 1.02 (0.96–1.09) | |
| ||||||||
NHS II | Person-years | 707,097 | 733,336 | 325,829 | 205,058 | 29,963 | 2,001,283 | |
Cases | 418 | 465 | 226 | 135 | 18 | 1,262 | ||
Crude Incidence/100000PY | 59 | 63 | 69 | 66 | 60 | |||
Age-adjusted model | 1 (Ref.) | 0.91 (0.79–1.05) | 1.02 (0.86–1.20) | 0.88 (0.72–1.08) | 0.78 (0.48–1.25) | 0.29 | 0.75 (0.50–1.11) | |
Multivariable-adjusted | 1 (Ref.) | 0.90 (0.79–1.04) | 1.04 (0.88–1.24) | 0.94 (0.76–1.15) | 0.85 (0.52–1.37) | 0.72 | 0.81 (0.54–1.22) | |
| ||||||||
HPFS | Person-years | 121,724 | 219,328 | 209,870 | 230,827 | 84,959 | 866,708 | |
Cases | 192 | 335 | 317 | 344 | 138 | 1,326 | ||
Crude Incidence/100000PY | 158 | 153 | 151 | 149 | 163 | |||
Age-adjusted model | 1 (Ref.) | 0.95 (0.79–1.13) | 0.99 (0.83–1.19) | 0.89 (0.74–1.07) | 0.86 (0.69–1.07) | 0.12 | 0.95 (0.81–1.10) | |
Multivariable-adjusted | 1 (Ref.) | 0.97 (0.80–1.16) | 1.02 (0.85–1.23) | 0.93 (0.77–1.12) | 0.91 (0.72–1.14) | 0.32 | 0.99 (0.85–1.16) | |
| ||||||||
Pooled* | Multivariable-adjusted | 1 (Ref.) | 0.95 (0.88–1.03) | 1.01 (0.92–1.10) | 0.95 (0.87–1.04) | 0.98 (0.86–1.13) | 0.88 | 1.02 (0.96–1.08) |
Abbreviations: NHS, Nurses’ Health Study; NHS II, Nurses’ Health Study II; HPFS, Health Professionals Follow-up Study. Multivariable analyses were adjusted for updated over time covariates: age (continuous); Caucasian (yes/no); body-mass index (<23, 23–24.9, 25–29.9, 30–34.9, ≥35 kg/m2); physical activity (metabolic-equivalents/week, quintiles); smoking status (never, past, current 1–14 cigarettes/d, current 15–24 cigarettes/d, current ≥25 cigarettes/d); physical examination for screening purposes (yes/no); current multivitamin use(yes/no); and current aspirin use(yes/no); family history of diabetes mellitus (yes/no), myocardial infarction (yes/no), or cancer(yes/no); history of diabetes mellitus (yes/no), hypertension (yes/no), or hypercholesterolemia (yes/no); intake of total energy, alcohol, red or processed meat, fruits, and vegetables (quintiles); and, in women, menopausal status and hormone use (premenopausal, postmenopausal never users, postmenopausal past users, postmenopausal current users). In NHS II, multivariable model was further adjusted for oral contraceptive use (never, past and current users). Frequency of nut consumption pertains to one serving of nuts, defined as 28 g. (1 oz)
Results from the multivariable model were combined with the use of the fixed-effects model.
In separate analyses of specific types of nuts (Central Illustration, Table 3), when comparing consumption of nuts two or more times per week with the reference category, the pooled multivariable-adjusted hazard ratios for CVD were 0.87 (95% CI, 0.82–0.93) for peanuts, and 0.85 (95% CI, 0.79–0.91) for tree nuts. Consuming walnuts one or more times per week was also associated with a lower risk of CVD [HR: 0.81 (95% CI, 0.71–0.91)]. Significant associations were also observed for each serving increase in peanuts, tree nuts and walnuts. Peanut butter intake was not significantly associated with CVD. Significant inverse associations were observed between the intake of total nuts, peanuts, tree nuts, and walnuts and CHD risk (Central Illustration, Online Table 5). Participants who consumed peanuts or tree nuts two or more times per week had respectively 15% (95% CI 8–21%) and 23% (95% CI 16–30%) lower risk of CHD compared to those who never or almost never consumed nuts. Consuming walnuts one or more times per week was associated with 21% (95% CI 6–34%) lower risk of CHD. Pooled analyses of the three cohorts showed that total nut consumption was inversely associated with fatal CHD [HR:0.69; 95% CI, 0.61–0.77, P<0.0001] but non-significant associations were observed for nonfatal myocardial infarction (Online Table 6).
Central Illustration. Cardiovascular Disease, Coronary Heart Disease and Stroke Based on Frequency of Nut Consumption and Type of Nut: Hazard Ratios.
Abbreviations: NHS, Nurses’ Health Study; NHS II, Nurses’ Health Study II; HPFS, Health Professionals Follow-up Study. Multivariable hazard ratios for total cardiovascular disease among study participants who consumed nuts two or more times per week (one or more times per week for walnuts) versus those who never or almost never consumed nuts, were adjusted for updated covariates: age (continuous); Caucasian (yes/no); body-mass index (<23, 23–24.9, 25–29.9, 30–34.9, ≥35 kg/m2); physical activity (metabolic-equivalents/week, quintiles); smoking status (never, past, current 1–14 cigarettes/d, current 15–24 cigarettes/d, current ≥25 cigarettes/d); physical examination for screening purposes (yes/no); current multivitamin use(yes/no); and current aspirin use(yes/no); family history of diabetes mellitus (yes/no), myocardial infarction (yes/no), or cancer(yes/no); history of diabetes mellitus (yes/no), hypertension (yes/no), or hypercholesterolemia (yes/no); intake of total energy, alcohol, red or processed meat, fruits, and vegetables (quintiles); and, in women, menopausal status and hormone use (premenopausal, postmenopausal never users, postmenopausal past users, postmenopausal current users). In NHS II, the multivariable model was further adjusted for oral contraceptive use (never, past and current users). Peanut butter models were additionally adjusted for glycemic load, soda, and white bread intake (quintiles). Results were pooled with the use of the fixed-effects model. Horizontal lines represent 95% confidence intervals.
Table 3.
Relative risk (95% CI) of cardiovascular events according to Type of Nuts
Frequency of Nut Consumption | ||||||
---|---|---|---|---|---|---|
| ||||||
Never or almost never | Less than once per Week | Once per Week | Two or more times per Week | P value for trend | HR (95% CI) per 28 g increase | |
Nurses’ Health Study | ||||||
| ||||||
Peanuts | ||||||
Median (g/d) | 0 | 1.30 | 3.92 | 12.04 | ||
Person-years | 999,858 | 829,970 | 204,015 | 164,431 | 2,198,274 | |
Cases | 3,158 | 2,544 | 567 | 458 | 6,727 | |
Crude Incidence/100000PY | 316 | 307 | 278 | 279 | ||
Multivariable-adjusted model | 1 (Ref.) | 0.92 (0.87–0.98) | 0.95 (0.87–1.04) | 0.90 (0.81–1.00) | 0.09 | 0.82 (0.68–0.99) |
Tree nuts | ||||||
Median (g/d) | 0 | 1.30 | 3.92 | 10.56 | ||
Person-years | 1,134,756 | 776,813 | 167,205 | 119,500 | 2,198,274 | |
Cases | 3,463 | 2,477 | 468 | 319 | 6,727 | |
Crude Incidence/100000PY | 305 | 319 | 280 | 267 | ||
Multivariable-adjusted model | 1 (Ref.) | 0.95 (0.90–1.00) | 0.95 (0.86–1.05) | 0.88 (0.78–0.99) | 0.03 | 0.79 (0.62–1.00) |
Walnuts | ||||||
Median (g/d) | 0 | 1.96 | 7.37 | |||
Person-years | 452,904 | 190,234 | 45,583 | 688,721 | ||
Cases | 1,880 | 654 | 106 | 2,640 | ||
Crude Incidence/100000PY | 422 | 348 | 237 | |||
Multivariable-adjusted model | 1 (Ref.) | 1.02 (0.93–1.12) | 0.72 (0.59–0.88) | 0.007 | 0.55 (0.34–0.89) | |
Peanut butter | ||||||
Median (g/d) | 0 | 0.90 | 2.10 | 6.30 | ||
Person-years | 402,929 | 728,603 | 405,168 | 661,574 | 2,198,274 | |
Cases | 1,028 | 2,358 | 1,284 | 2,057 | 6,727 | |
Crude Incidence/100000PY | 255 | 324 | 317 | 311 | ||
Multivariable-adjusted model | 1 (Ref.) | 1.06 (0.98–1.14) | 1.06 (0.97–1.15) | 1.04 (0.96–1.13) | 0.75 | 0.95 (0.87–1.04) |
| ||||||
Nurses’ Health Study II | ||||||
| ||||||
Peanuts | ||||||
Median (g/d) | 0 | 1.17 | 3.92 | 8.02 | ||
Person-years | 909,678 | 876,295 | 139,660 | 75,077 | 2,000,710 | |
Cases | 837 | 869 | 135 | 74 | 1,915 | |
Crude Incidence/100000PY | 92 | 99 | 97 | 99 | ||
Multivariable-adjusted model | 1 (Ref.) | 0.90 (0.81–0.99) | 0.88 (0.73–1.07) | 0.83 (0.65–1.06) | 0.09 | 0.55 (0.31–1.00) |
Tree nuts | ||||||
Median (g/d) | 0 | 1.30 | 3.92 | 8.79 | ||
Person-years | 990,483 | 730,724 | 152,825 | 126,678 | 2,000,710 | |
Cases | 891 | 750 | 157 | 117 | 1,915 | |
Crude Incidence/100000PY | 90 | 103 | 103 | 92 | ||
Multivariable-adjusted model | 1 (Ref.) | 0.93 (0.83–1.03) | 0.92 (0.77–1.10) | 0.80 (0.65–0.98) | 0.04 | 0.53 (0.33–0.85) |
Walnuts | ||||||
Median (g/d) | 0 | 0.68 | 4.65 | |||
Person-years | 588,362 | 443,135 | 63,826 | 1,095,323 | ||
Cases | 732 | 458 | 52 | 1,242 | ||
Crude Incidence/100000PY | 124 | 103 | 81 | |||
Multivariable-adjusted model | 1 (Ref.) | 0.84 (0.74–0.95) | 0.66 (0.50–0.89) | 0.003 | 0.21 (0.06–0.73) | |
Peanut butter | ||||||
Median (g/d) | 0 | 1.05 | 2.10 | 6.10 | ||
Person-years | 351,026 | 701,495 | 400,266 | 547,923 | 2,000,710 | |
Cases | 390 | 661 | 397 | 467 | 1,915 | |
Crude Incidence/100000PY | 111 | 94 | 99 | 85 | ||
Multivariable-adjusted model | 1 (Ref.) | 0.77 (0.67–0.87) | 0.87 (0.75–1.00) | 0.73 (0.63–0.84) | 0.004 | 0.84 (0.68–1.04) |
| ||||||
Health Professionals’ Follow-up Study | ||||||
| ||||||
Peanuts | ||||||
Median (g/d) | 0 | 1.96 | 3.92 | 12.04 | ||
Person-years | 177,482 | 370,105 | 145,053 | 171,815 | 864,455 | |
Cases | 1,288 | 2,234 | 875 | 1,097 | 5,494 | |
Crude Incidence/100000PY | 726 | 604 | 603 | 638 | ||
Multivariable-adjusted model | 1 (Ref.) | 0.91 (0.85–0.98) | 0.93 (0.85–1.02) | 0.85 (0.78–0.93) | 0.002 | 0.95 (0.85–1.06) |
Tree nuts | ||||||
Median (g/d) | 0 | 1.63 | 3.92 | 11.01 | ||
Person-years | 257,997 | 377,087 | 118,784 | 110,587 | 864,455 | |
Cases | 1,818 | 2,330 | 728 | 618 | 5,494 | |
Crude Incidence/100000PY | 705 | 618 | 613 | 559 | ||
Multivariable-adjusted model | 1 (Ref.) | 0.94 (0.88–1.00) | 0.98 (0.90–1.07) | 0.84 (0.76–0.92) | 0.001 | 0.85 (0.72–1.00) |
Walnuts | ||||||
Median (g/d) | 0 | 1.30 | 6.02 | |||
Person-years | 142,208 | 65,094 | 24,124 | 231,426 | ||
Crude Incidence/100000PY | ||||||
Cases | 895 | 349 | 129 | 1,373 | ||
Multivariable-adjusted model | 1 (Ref.) | 0.95 (0.83–1.08) | 0.99 (0.81–1.20) | 0.84 | 1.01 (0.65–1.56) | |
Peanut butter | ||||||
Median (g/d) | 0 | 1.05 | 2.10 | 6.45 | ||
Person-years | 227,180 | 233,480 | 134,685 | 269,109 | 864,454 | |
Cases | 1,454 | 1,431 | 832 | 1,777 | 5,494 | |
Crude Incidence/100000PY | 705 | 618 | 613 | 559 | ||
Multivariable-adjusted model | 1 (Ref.) | 0.97 (0.90–1.05) | 1.02 (0.93–1.11) | 1.01 (0.93–1.08) | 0.55 | 1.03 (0.96–1.10) |
| ||||||
Pooled* | ||||||
| ||||||
Peanuts | ||||||
Multivariable-adjusted model | 1 (Ref.) | 0.92 (0.88–0.95) | 0.94 (0.88–1.00) | 0.87 (0.82–0.93) | 0.0002 | 0.91 (0.83–1.00) |
Tree nuts | ||||||
Multivariable-adjusted model | 1 (Ref.) | 0.95 (0.91–0.98) | 0.96 (0.90–1.03) | 0.85 (0.79–0.91) | 0.002 | 0.81 (0.71–0.92) |
Walnuts | ||||||
Multivariable-adjusted model | 1 (Ref.) | 0.95 (0.89–1.02) | 0.81 (0.71–0.92) | <0.001 | 0.71 (0.52–0.97) | |
Peanut butter | ||||||
Multivariable-adjusted model | 1 (Ref.) | 0.98 (0.93–1.02) | 1.01 (0.96–1.07) | 0.99 (0.94–1.04) | 0.74 | 0.99 (0.94–1.05) |
Multivariable analyses were adjusted for updated over time covariates: age (continuous); Caucasian (yes/no); body-mass index (<23, 23–24.9, 25–29.9, 30–34.9, ≥35 kg/m2); physical activity (metabolic-equivalents/week, quintiles); smoking status (never, past, current 1–14 cigarettes/d, current 15–24 cigarettes/d, current ≥25 cigarettes/d); physical examination for screening purposes (yes/no); current multivitamin use(yes/no); and current aspirin use(yes/no); family history of diabetes mellitus (yes/no), myocardial infarction (yes/no), or cancer(yes/no); history of diabetes mellitus (yes/no), hypertension (yes/no), or hypercholesterolemia (yes/no); intake of total energy, alcohol, red or processed meat, fruits, and vegetables (quintiles); and, in women, menopausal status and hormone use (premenopausal, postmenopausal never users, postmenopausal past users, postmenopausal current users). In NHS II, multivariable model was further adjusted for oral contraceptive (never, past and current users). Peanut butter models were additionally adjusted for glycaemic load, soda and white bread intake (quintiles). Frequency of nut consumption pertains to one serving of nuts, defined as 28 g.
Results from the multivariable model were combined with the use of the fixed-effects model.
The pooled risk estimates for stroke among participants who consumed total nuts and peanuts two or more times per week was 0.94 (95% CI, 0.88–1.05, P=0.08) and 0.90 (95% CI, 0.81–0.99, P=0.07), respectively. Walnut intake was associated with a 17% (95% CI, 4%–29%, P=0.10) lower risk of stroke (Central Illustration, Online Table 5). Peanut butter and tree nuts were not associated with stroke risk (Online Table 5). Likewise, we found no significant associations between total nut consumption and the risk of fatal, nonfatal or ischemic stroke (Online Table 7).
In analyses stratified by potential risk factors for CVD, the inverse association between total nut consumption and CVD persisted in most subgroups (Figure 1). No significant interactions were observed in pooled analysis. Our results remained robust in several sensitivity analyses. The results remained unchanged when we continuously updated diet until the end of follow-up (pooled multivariable HR for CVD comparing nut consumption five or more times per week with no nut consumption: HR: 0.86, 95% CI, 0.78–0.94, P<0.001) and when we used the most recent diet as our primary exposure (pooled multivariable HR for CVD comparing nut consumption five or more times per week with no nut consumption: HR: 0.88 (95% CI, 0.82–0.94 P<0.001). Additional adjustment for tooth loss or AHEI (excluding nut component) did not materially alter the associations (data not shown). When specific types of nuts were mutually adjusted for other nuts in individual cohorts, results were attenuated (i.e. peanuts mutually adjusted for tree nuts and peanut butter). However, significant inverse associations remained in the pooled analysis for peanuts and tree nuts (Online Table 8). Significant inverse associations between baseline nut consumption and risk of CVD and CHD were also observed (Online Table 9). Results were consistent with primary analysis when we excluded BMI from the multivariable adjusted model (pooled multivariable HR for CVD comparing nut consumption five or more times per week with no nut consumption: HR: 0.85 (95% CI, 0.79–0.93 P<0.001).
Figure 1. Hazard Ratios for Cardiovascular Disease in Subgroups.
Multivariable hazard ratios for pooled analysis of the three cohorts (Nurses’ Health Study I and II and Health Professionals Follow-up Study) for total cardiovascular disease among study participants who consumed nuts two or more times per week versus those who never consumed nuts were adjusted for updated covariates: age (continuous); Caucasian (yes/no); body-mass index (<23, 23–24.9, 25–29.9, 30–34.9, ≥35 kg/m2); physical activity (metabolic-equivalents/week, quintiles); smoking status (never, past, current 1–14 cigarettes/d, current 15–24 cigarettes/d, current ≥25 cigarettes/d); physical examination for screening purposes (yes/no); current multivitamin use(yes/no); and current aspirin use(yes/no); family history of diabetes mellitus (yes/no), myocardial infarction (yes/no), or cancer(yes/no); history of diabetes mellitus (yes/no), hypertension (yes/no), or hypercholesterolemia (yes/no); intake of total energy, alcohol, red or processed meat, fruits, and vegetables (quintiles); and, in women, menopausal status and hormone use (premenopausal, postmenopausal never users, postmenopausal past users, postmenopausal current users). In NHS II, the multivariable model was further adjusted for oral contraceptive (never, past and current users). Results were pooled with the use of the fixed-effects model. Horizontal lines represent 95% confidence intervals.
Discussion
In three large prospective cohorts with up to 32 years of follow-up, we observed that nut consumption was associated with lower risk of developing CVD after adjusting for cardiovascular risk factors. As compared to those participants who never or almost never consume nuts, those who consumed nuts five or more times per week had 14% lower risk of CVD and 20% lower risk of CHD. Results were similar for tree nuts, peanuts, and walnuts; and the inverse association persisted across all subgroups in stratified analysis. Significant inverse associations were observed for each increasing serving of nuts and risk of CVD and CHD. Although we found no evidence of an association between total nut consumption and risk of stroke, the intake of peanuts and walnuts was inversely associated with the risk of stroke.
Our results are in line with previous observational studies that reported inverse associations between nut consumption and CVD risk. However, the number of studies that investigated specific types of nuts is limited and few of them had repeated measures of diet (11, 12). To date, twelve cohort studies have investigated the association between total nut consumption and CVD risk (11). Consistent with our findings, a recent meta-analysis of these twelve studies found a summary relative risk for high versus low intake of nuts of 0.81 (95 % CI: 0.74–0.89) for CVD and 0.76 (95% CI: 0.69–0.84) for CHD (11). Our results are in agreement with a meta-analysis of 7 prospective studies that showed a 30% lower risk of CHD mortality in participants with higher nut consumption and no significant associations for non-fatal CHD, but only 3 studies examined nut consumption in relation to non-fatal CHD (12).
Despite studies reporting consistent inverse associations between higher nut consumption and risk of CVD and CHD, studies evaluating associations with stroke remain limited (11). In the pooled analysis of our cohorts, we found that consumption of peanuts two or more times per week and walnuts one or more times per week was associated with a 10% and 17% lower risk of total stroke, respectively, but there was no significant associations for total nut intake. In a meta-analysis of eleven prospective studies, an inverse association was observed for total nut consumption and total stroke risk [HR: 0.89 (95% CI: 0.82–0.97) for high vs. low intake] (11). However, the associations were not significant in eight of the individual studies included, and most of the studies did not differentiate between fatal and nonfatal stroke. Two studies evaluated the associations between intake of specific types of nuts and stroke death. The Shanghai Men’s Health Study (23) reported an inverse trend between higher peanut intake and lower risk of ischemic stroke (HR comparing 5th versus 1st quintile = 0.77 (0.60–1.00), P-trend=0.003). In the Netherlands Cohort Study (24), an intake of more than 5 g/d of peanuts was associated with a 29% (6–46%) lower risk of stroke death compared to those who did not consume peanuts. Another reason that may account for the lack of associations between nut intake and stroke, is the lack of effect of nuts on blood pressure (25).
Walnuts are among the most widely consumed tree nut in the world and are high in n-6 and n-3 PUFA (especially in plant-derived α-linolenic acid). Their consumption has been associated with cardio protective properties (26). Although numerous clinical studies have shown beneficial effects of walnuts on lipid profiles and inflammatory biomarkers (27), evidence from large prospective studies with long durations of follow-up is sparse. In a secondary analysis of the PREDIMED trial, participants who consumed >3 servings per week of walnuts had 47% (2% to 71%) lower risk of cardiovascular mortality compared to those who did not consume walnuts (28). We have also observed that consuming walnuts at least once per week was associated with 19% lower risk of CVD, 21% lower risk of CHD and 17% lower risk of stroke.
Peanut butter intake was inversely associated with CVD and stroke in the NHS II but no significant associations were observed when pooling the three cohort studies together. Although we conducted several sensitivity analyses adjusting peanut butter models for foods associated with peanut butter consumption (such as soda, white bread, and glycaemic load), residual confounding by dietary pattern remains a strong possibility. Further, it may be possible that peanut butter intake is not associated with disease risk. In a previous report of the NHS and HPFS, peanut butter intake was not associated with inflammatory biomarkers (29). Of note, in our cohorts, a serving of peanut butter was defined as 1 tablespoon (15 g), which is not consistent with the typical serving size of 2 tablespoons and may have misclassified intake levels. Further studies are needed to elucidate the health effects of peanut butter intake.
Findings from observational studies alone cannot be used to draw conclusions regarding whether associations are causal, however, the PREDIMED trial found that participants who were randomized to a Mediterranean diet supplemented with mixed nuts had 28% (95% 0.54–0.96) reduced risk of a composite of cardiovascular events and 46% (95% CI, 0.35–0.84) reduced risk of stroke after 4.8 years of follow-up compared to a control diet (advice to reduce all types of dietary fat) (9).
Nuts are high in unsaturated fatty acids, dietary fiber, minerals, vitamins and several bioactive compounds, which may in part explain their beneficial effects on cardiovascular health (30). There are several mechanisms that may account for the inverse associations between nut consumption and CVD. Randomized controlled trials have shown that consumption of nuts improves lipid profiles (5, 27), attenuate inflammation (31), oxidative stress and endothelial function (32), and decreases insulin resistance (33). Nuts are also rich in polymerized polyphenols which provide a substrate for gut microbiota. The compounds arising from this metabolism may modulate gut microbiota through prebiotic effects and antimicrobial activities and consequently contribute to cardiovascular benefits for the host (34). Despite nuts being an energy-dense food, there is no scientific evidence supporting associations between weight gain and nut consumption. Indeed, they have been associated with lower weight gain and lower risk of obesity, probably because they can increase satiety and fullness which may potentially reduce the consumption of unhealthy snacks (35–37).
The strengths of the present study include its prospective design, large sample size including men and women, long duration of follow-up with a high retention rate, repeated assessment of diet and lifestyle variables, and analyses of several CVD outcomes including fatal and nonfatal CVD, fatal and nonfatal CHD, and stroke (fatal, nonfatal and ischemic). Moreover, we present strong evidence for the association between total nuts, peanuts, tree nuts, walnuts and peanut butter with CVD risk in three large cohort studies with more than 13,500 CVD cases. Several limitations deserve comment. First, given that our study sample was limited to white health professionals, this could limit generalizability of our findings. Still, because there is no reason to expect that the underlying biological mechanisms may be different, our results can be generalized to men and women of different ethnicities. Because nut intake was self-reported, some measurement error is inevitable. However, because dietary data were collected prospectively, misreporting is likely to be random, resulting in an underestimation of the association. By calculating the cumulative average of nut intake from multiple time points, we were able to reduce potential random measurement error. Further, inverse associations were consistent across pre-specified subgroup analysis and in several sensitivity analyses. Because we lacked data on how nuts were prepared (e.g., salted, raw, roasted), we were unable to test the influence of preparation methods.
In conclusion, findings from three large prospective cohort studies indicate that frequent intake of nuts, tree nuts, peanuts and walnuts was associated with a lower risk of CVD, independently from other cardiovascular risk, lifestyle, and dietary factors. Our findings support recommendations of increasing the intake of a variety of nuts, as part of healthy dietary patterns, to reduce the risk of chronic diseases in the general population.
Supplementary Material
Perspectives.
Competency in Medical Knowledge
Consumption of peanuts and tree nuts two or more times per week and walnuts one or more times/week is associated with a lower risk of coronary artery disease and cardiovascular disease in general.
Competency in Patient Care
Increasing intake of nuts, as part of a healthy diet, may help to reduce the risk of cardiovascular disease in the general population.
Translational Outlook
Further research is needed to investigate the mechanisms underlying the association between nut consumption and reduction in cardiovascular risk.
Acknowledgments
Funding: This study was supported by research grants UM1 CA186107, R01 HL034594, R01 HL088521, UM1 CA176726, UM1 CA167552, R01 HL35464, R01 HL60712 from the National Institutes of Health, United States. SNB is supported by a Career Development Grant from the National Institutes of Health (K01 DK107804).
We thank the participants and staff of the Nurses’ Health Study, Nurses’ Health Study II, and Health Professionals Follow-up Study for their valuable contributions.
Abbreviations
- CHD
coronary heart disease
- CI
confidence interval
- CVD
cardiovascular disease
- FFQ
food frequency
- HPFS
Health Professionals Follow-up study
- HR
hazard ratio
- NHS
Nurses’ Health Study
- US
United States
Footnotes
Disclosures: Drs. Yanping Li and Frank Hu have received research support from California Walnut Commission. Dr. Vasanti Malik has received research support from the Peanut Institute. Neither the corresponding author nor other authors have any conflict of interest affecting the conduct or reporting of the work submitted.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.GBD 2013 Mortality and Causes of Death Collaborators. Global, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2015;385:117–71. doi: 10.1016/S0140-6736(14)61682-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.U.S. Department of Agriculture and U.S. Department of Helath and Human Services. Scientific Report of the 2015 Dietary Guidelines Advisory Committee. 2015 [Google Scholar]
- 3.Ros E. Health benefits of nut consumption. Nutrients. 2010;2:652–682. doi: 10.3390/nu2070652. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Ros E, Hu FB. Consumption of plant seeds and cardiovascular health: epidemiological and clinical trial evidence. Circulation. 2013;128:553–65. doi: 10.1161/CIRCULATIONAHA.112.001119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Sabate J, Oda K, Ros E. Nut consumption and blood lipid levels: a pooled analysis of 25 intervention trials. Arch Intern Med. 2010;170:821–827. doi: 10.1001/archinternmed.2010.79. [DOI] [PubMed] [Google Scholar]
- 6.Kendall CW, Josse AR, Esfahani A, Jenkins DJ. Nuts, metabolic syndrome and diabetes. Br J Nutr. 2010;104:465–473. doi: 10.1017/S0007114510001546. [DOI] [PubMed] [Google Scholar]
- 7.Luo C, Zhang Y, Ding Y, et al. Nut consumption and risk of type 2 diabetes, cardiovascular disease, and all-cause mortality: a systematic review and meta-analysis. Am J Clin Nutr. 2014;100:256–269. doi: 10.3945/ajcn.113.076109. [DOI] [PubMed] [Google Scholar]
- 8.Hu FB, Manson JE, Willett WC. Types of dietary fat and risk of coronary heart disease: a critical review. J Am Coll Nutr. 2001;20:5–19. doi: 10.1080/07315724.2001.10719008. [DOI] [PubMed] [Google Scholar]
- 9.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. doi: 10.1056/NEJMc1806491. [DOI] [PubMed] [Google Scholar]
- 10.Bao Y, Han J, Hu FB, et al. Association of nut consumption with total and cause-specific mortality. N Engl J Med. 2013;369:2001–11. doi: 10.1056/NEJMoa1307352. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Aune D, Keum N, Giovannucci E, et al. Nut consumption and risk of cardiovascular disease, total cancer, all-cause and cause-specific mortality: a systematic review and dose-response meta-analysis of prospective studies. BMC Med. 2016;14:207. doi: 10.1186/s12916-016-0730-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mayhew AJ, de Souza RJ, Meyre D, Anand SS, Mente A. A systematic review and meta-analysis of nut consumption and incident risk of CVD and all-cause mortality. Br J Nutr. 2016;115:212–25. doi: 10.1017/S0007114515004316. [DOI] [PubMed] [Google Scholar]
- 13.Belanger C, Speizer FE, Hennekens CH, Rosner B, Willett W, Bain C. The nurses’ health study: current findings. Am J Nurs. 1980;80:1333. doi: 10.1097/00000446-198007000-00024. [DOI] [PubMed] [Google Scholar]
- 14.Colditz GA, Martin P, Stampfer MJ, et al. Validation of questionnaire information on risk factors and disease outcomes in a prospective cohort study of women. Am J Epidemiol. 1986;123:894–900. doi: 10.1093/oxfordjournals.aje.a114319. [DOI] [PubMed] [Google Scholar]
- 15.Rimm EB, Giovannucci EL, Stampfer MJ, Colditz GA, Litin LB, Willett WC. Reproducibility and validity of an expanded self-administered semiquantitative food frequency questionnaire among male health professionals. Am J Epidemiol. 1992;135:1114-26-36. doi: 10.1093/oxfordjournals.aje.a116211. [DOI] [PubMed] [Google Scholar]
- 16.Mendis S, Thygesen K, Kuulasmaa K, et al. World Health Organization definition of myocardial infarction: 2008–09 revision. Int J Epidemiol. 2011;40:139–46. doi: 10.1093/ije/dyq165. [DOI] [PubMed] [Google Scholar]
- 17.Walker AE, Robins M, Weinfeld FD. The National Survey of Stroke. Clinical findings. Stroke. 12:I13–44. [PubMed] [Google Scholar]
- 18.Stampfer MJ, Willett WC, Speizer FE, et al. Test of the National Death Index. Am J Epidemiol. 1984;119:837–9. doi: 10.1093/oxfordjournals.aje.a113804. [DOI] [PubMed] [Google Scholar]
- 19.Salvini S, Hunter DJ, Sampson L, et al. Food-based validation of a dietary questionnaire: the effects of week-to-week variation in food consumption. Int J Epidemiol. 1989;18:858–67. doi: 10.1093/ije/18.4.858. [DOI] [PubMed] [Google Scholar]
- 20.Howell TH, Ridker PM, Ajani UA, Hennekens CH, Christen WG. Periodontal disease and risk of subsequent cardiovascular disease in U.S. male physicians. J Am Coll Cardiol. 2001;37:445–50. doi: 10.1016/s0735-1097(00)01130-x. [DOI] [PubMed] [Google Scholar]
- 21.McCullough ML, Willett WC. Evaluating adherence to recommended diets in adults: the Alternate Healthy Eating Index. Public Health Nutr. 2006;9:152–7. doi: 10.1079/phn2005938. [DOI] [PubMed] [Google Scholar]
- 22.Luu HN, Blot WJ, Xiang Y-B, et al. Prospective evaluation of the association of nut/peanut consumption with total and cause-specific mortality. JAMA Intern Med. 2015;175:755–66. doi: 10.1001/jamainternmed.2014.8347. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.van den Brandt PA, Schouten LJ. Relationship of tree nut, peanut and peanut butter intake with total and cause-specific mortality: a cohort study and meta-analysis. Int J Epidemiol. 2015;44:1038–49. doi: 10.1093/ije/dyv039. [DOI] [PubMed] [Google Scholar]
- 24.Mohammadifard N, Salehi-Abargouei A, Salas-Salvadó J, Guasch-Ferré M, Humphries K, Sarrafzadegan N. The effect of tree nut, peanut, and soy nut consumption on blood pressure: a systematic review and meta-analysis of randomized controlled clinical trials. Am J Clin Nutr. 2015;101:966–82. doi: 10.3945/ajcn.114.091595. [DOI] [PubMed] [Google Scholar]
- 25.Kris-Etherton PM. Walnuts Decrease Risk of Cardiovascular Disease: A Summary of Efficacy and Biologic Mechanisms. J Nutr. 2014;144:547S–554S. doi: 10.3945/jn.113.182907. [DOI] [PubMed] [Google Scholar]
- 26.Banel DK, Hu FB. Effects of walnut consumption on blood lipids and other cardiovascular risk factors: a meta-analysis and systematic review. Am J Clin Nutr. 2009;90:56–63. doi: 10.3945/ajcn.2009.27457. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Guasch-Ferre M, Bullo M, Martinez-Gonzalez MA, et al. Frequency of nut consumption and mortality risk in the PREDIMED nutrition intervention trial. BMC Med. 2013;11:164. doi: 10.1186/1741-7015-11-164. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Yu Z, Malik VS, Keum N, et al. Associations between nut consumption and inflammatory biomarkers. Am J Clin Nutr. 2016;104:722–728. doi: 10.3945/ajcn.116.134205. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Ros E, Tapsell LC, Sabate J. Nuts and berries for heart health. Curr Atheroscler Rep. 2010;12:397–406. doi: 10.1007/s11883-010-0132-5. [DOI] [PubMed] [Google Scholar]
- 30.Salas-Salvado J, Garcia-Arellano A, Estruch R, et al. Components of the Mediterranean-type food pattern and serum inflammatory markers among patients at high risk for cardiovascular disease. Eur J Clin Nutr. 2008;62:651–659. doi: 10.1038/sj.ejcn.1602762. [DOI] [PubMed] [Google Scholar]
- 31.Bulló M, Lamuela-Raventós R, Salas-Salvadó J. Mediterranean diet and oxidation: nuts and olive oil as important sources of fat and antioxidants. Curr Top Med Chem. 2011;11:1797–1810. doi: 10.2174/156802611796235062. [DOI] [PubMed] [Google Scholar]
- 32.Casas-Agustench P, Bullo M, Salas-Salvado J. Nuts, inflammation and insulin resistance. Asia Pac J Clin Nutr. 2010;19:124–130. [PubMed] [Google Scholar]
- 33.Lamuela-Raventos RM, St Onge M-P. Prebiotic Nut Compounds and Human Microbiota. Crit Rev Food Sci Nutr. 2016:00–00. doi: 10.1080/10408398.2015.1096763. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Bes-Rastrollo M, Sabate J, Gomez-Gracia E, Alonso A, Martinez JA, Martinez-Gonzalez MA. Nut consumption and weight gain in a Mediterranean cohort: The SUN study. Obesity (Silver Spring) 2007;15:107–116. doi: 10.1038/oby.2007.507. [DOI] [PubMed] [Google Scholar]
- 35.Bes-Rastrollo M, Wedick NM, Martinez-Gonzalez MA, Li TY, Sampson L, Hu FB. Prospective study of nut consumption, long-term weight change, and obesity risk in women. Am J Clin Nutr. 2009;89:1913–1919. doi: 10.3945/ajcn.2008.27276. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Freisling H, Noh H, Slimani N, et al. Nut intake and 5-year changes in body weight and obesity risk in adults: results from the EPIC-PANACEA study. Eur J Nutr. 2017 doi: 10.1007/s00394-017-1513-0. In press. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.