Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Eur J Clin Nutr. Author manuscript; available in PMC 2011 January 1.
Published in final edited form as:
PMCID: PMC2802656

Nut Consumption and Risk of Type 2 Diabetes in the Physicians' Health Study



While type 2 Diabetes (DM) is a major cause of morbidity in the US, limited data are available on the association between nut intake and incident DM. We sought to examine the association between nut consumption and the risk of DM.


Prospective cohort of 20,224 male participants of the Physicians' Health Study I. Consumption of nuts was estimated using a 19-item brief food frequency questionnaire and incident DM was ascertained through yearly follow-up questionnaires. Cox regression was used to estimate relative risks of DM.


The average age was 54.4 ± 9.4 years (range 40.7–87.1). During a mean follow up of 19.2 years, 1,828 cases of DM occurred. The crude incidence rates of DM were 4.82, 4.85, 4.92, 4.16, 4.29, and 3.32 cases/1,000 person-years for people reporting nut consumption of rarely/never, <1, 1, 2–4, 5–6, and 7+ servings/week, respectively. While nut consumption was associated with a lower risk of DM in a model adjusted for age (p for tend 0.017), such relation was attenuated upon additional control for other confounders (multivariable adjusted hazard ratios (HR) (95% CI) for DM were 1.0 (reference), 1.06 (0.93–1.20), 1.10 (0.95–1.26), 0.97 (0.82–1.14), 0.99 (0.76–1.30), and 0.87 (0.61–1.24) from the lowest to the highest category of nut consumption, respectively (p for trend 0.99). No statistically significant association between nut consumption and DM was found in either lean or overweight/obese participants.


Our data do not show an association between nut consumption and incident DM in US male physicians.

Keywords: Nuts, type 2 diabetes, epidemiology, risk factors


Diabetes affects 23.6 million Americans, and accounted for an estimated $174 billion in direct and indirect costs in 2007 in the United States (NIDDK 2008). As a leading cause of blindness and renal failure, it adds to a significant morbidity and mortality burden on healthcare system. Several prospective studies have suggested an inverse association between consumption of dietary fiber, vegetables, fish, moderate amount of alcohol, whole grain products and incident DM (Parillo and Riccardi 2004, van Dam 2003). Similarly, in the Physicians' Health Study (PHS) I, it has been shown that consumption of breakfast cereals (Kochar et al 2007) and avoidance of egg consumption (Djousse et al 2009a) are associated with decreased risk of DM. Although a wealth of literature has shown various health benefits of nuts in dyslipidemia in established diabetics (Lovejoy et al 2002), glycemic control (Josse et al 2007) and reduced oxidative damage (Jenkins et al 2006), only one study has suggested that tree nut and peanut butter consumption may be associated with reduced incidence of DM in women (Jiang et al 2002). Furthermore, nut consumption has been inversely associated with markers of inflammation, another risk factor for DM (Jiang et al 2006).

The annual per capita tree nut and peanut availability in the US in 2006 was about 10 lbs, and there has been an upward trend in this since 1965 (USDA 2008). Despite its growing popularity as a health promoting food, sparse data are available on the effect of nuts on the risk of DM. We sought to explore prospectively the association between nut consumption and incident DM among 20,224 US male physicians.

Subjects and methods

Study population

We used data from PHS I, a completed randomized, double blind, placebo-controlled trial designed to study low-dose aspirin and β-carotene for the primary prevention of cardiovascular disease and cancer. A detailed description of the PHS I was previously published (Steering Committee of the Physicians' Health Study Research Group 1989). Of the total 22 071 participants, we excluded 615 subjects because of missing data on nut consumption, 684 subjects with prevalent DM at the time of the exposure assessment, 68 cases who died before 12 months after randomization or whose last questionnaire was received before 12 months after randomization, and 480 subjects with missing covariates. Thus, a final sample of 20,224 participants was used for the current analyses. Each participant signed an informed consent, and the Institutional Review Board at Brigham and Women's Hospital approved the study protocol.

Nut consumption

We used a 19-item brief food frequency questionnaire to obtain self-reported information on nut consumption at 12 months after randomization (1983–1985). Participants were asked to report how often, on average, they have consumed nuts (small packet of 1 oz) during the past year (possible responses were rarely/never, 1–3 servings/month, 1 serving/wk, 2–4 servings/wk, 5–6 servings/wk, daily consumption, and 2 servings/d). Although the food frequency questionnaire was not validated in this cohort, it has been validated in several cohorts (Rimm et al 1992, Salvini et al 1989, Stein et al 1992, Willett et al 1985). We did not query about peanuts as a separate question; thus, we are unable to make a distinction between peanuts and tree nuts.

Ascertainment of incident DM

A questionnaire was mailed to each participant every 6 months during the first year and has been mailed annually thereafter to obtain information on incident outcomes including DM.

Other variables

Information on hypertension was collected through self-reported annual follow-up questionnaires. Data on demographics, anthropometrics, history of hypertension (defined as self-reported systolic blood pressure (BP) ≥140 mm Hg, diastolic BP ≥90 mm Hg, or use of antihypertensive medications), intake of selected foods such as fruit and vegetables, red meat, consumption of breakfast cereal, physical activity, smoking, and alcohol consumption were obtained at baseline (1982– 1983). Physical activity was ascertained at baseline with a single question, ”How often do you exercise vigorously enough to work up a sweat?”. For the purpose of current study, they were dichotomized into 1 or more times per week or < 1 per week. The rationale of inclusion of hypertension in the model is that inadequately controlled blood pressure in non-diabetics has been shown to be associated with an increased risk of incident DM (Izzo et al 2009). Questions on the type of dairy products, i.e. milk, yogurt and cheese were asked, but data on fat content of these products, i.e. low fat, fat free products were not available.

Statistical analyses

We classified each subject into one of the following categories of nut consumption: Rarely/never, <1, 1, 2–4, 5–6, and 7+ servings/week. We computed person-time of follow-up from exposure assessment (12 months after randomization) until the first occurrence of 1) DM, 2) death, or 3) date of receipt of last follow-up questionnaire. We used Cox proportional hazard models to compute multivariable-adjusted HR with corresponding 95% CIs with the use of subjects in the lowest category of nut consumption as the reference group. We assessed confounding by established risk factors for DM. The initial model only adjusted for age, and subsequent models adjusted for age and exercise, age and hypertension, age and BMI, and a final model controlled for age, smoking (never, past and current smokers), aspirin assignment, breakfast cereal consumption (0, ≤1, 2–6, and 7+servings/week), dairy consumption (quintiles),and red meat consumption (quintiles).

Assumptions for proportional hazard models were tested (by including main effects and product terms of covariates and logarithmic-transformed time factor) and were met (all P values > 0.05). In a secondary analysis, we examined whether adiposity modified the association between nut intake and DM with BMI of 25 kg/m2 as the cutoff to separate lean from overweight or obese subjects. We then conducted stratified analyses by adiposity status (BMI < 25 or 25+ kg/m2) and tested statistical interaction with the use of a product term of nut consumption and adiposity variable in a hierarchical model. All analyses were completed with the use of SAS, version 9.1 (SAS Institute Inc, Cary, NC, USA). Significance level was set at 0.05.


The mean age was 54.4 ± 9.4 years (range 40.7–87.2) among the 20,224 male participants. Higher nut consumption was associated with higher breakfast cereal intake, and lower prevalence of hypertension (Table 1). The crude incidence rates of DM were 4.82, 4.85, 4.92, 4.16, 4.29, and 3.32 cases/1,000 person-years for people reporting rarely/never, <1, 1, 2–4, 5–6, and 7+ servings/week of nut consumption, respectively. From the lowest to the highest category of nut consumption, age- adjusted HR (95% CI) for DM were 1.0 (reference), 1.01 (0.89–1.15), 1.03 (0.90–1.18), 0.87 (0.73–1.02), 0.88 (0.67–1.16), and 0.68 (0.48–0.97), respectively (P for trend 0.017, Table 2). However, additional adjustment for either exercise (HR: 7+ servings of nuts /week vs. none = 0.71 (0.50–1.02), body mass index (corresponding HR: 0.77 (0.54–1.11), or prevalent hypertension (HR: 0.74 (0.52–1.06) led to the attenuation of the inverse association between nuts and DM. In a multivariable model with adjustment for age, randomization's arm, cigarette smoking, body mass index, history of hypertension, physical activity, alcohol intake, and dairy, red meat, breakfast cereal consumption, a statistically non-significant 13% lower risk of incident DM was observed in those consuming 7+ servings of nuts/week, as compared to those who rarely or never ate nuts (p for linear trend 0.99, Table 2). Similarly, no statistically significant associations were found between nut consumption and DM in subgroup analyses of lean and overweight or obese people (Table 3).

Table 1
Characteristics of 20,224 participants according to categories of nut consumption in the Physicians' Health Study*
Table 2
Hazard ratios (95% CI) for type 2 diabetes according to nut consumption in the Physicians' Health Study *
Table 3
Hazard ratios (95% CI) of DM according to nut consumption and body mass index (BMI) in the Physicians' Health Study *


In this study, we found that nut consumption is not associated with incident DM in apparently healthy US male physicians. The observed inverse relation between nuts and DM observed in the age-adjusted model was mainly attenuated by additional adjustment of exercise, body mass index, or prevalent hypertension. Being lean, overweight or obese did not modify these findings. To the best of our knowledge, this is the first large prospective study to evaluate effect of nut consumption on incident DM in a male population.

Jiang et al. (Jiang et al 2002) have shown in a large prospective study of 83,818 women in the Nurses' Health Study that nut consumption was associated with a 29% decreased risk of incident DM. In women consuming nuts ≥5 times per week, a 45% risk reduction was seen in those with BMI<25 kg/m2. Residual confounding by body weight was considered as a major concern by the authors. As female bodies have a different hormonal milieu than their male counterparts, it is difficult to compare our results with that study. Moreover, in a letter to the editor in response to that study, researchers from the Iowa Women's Health Study presented an analysis, which showed a positive association between nut consumption and risk of incident DM (Parker et al 2003). Similarly, many interventional studies in diabetics have shown no effect of nuts on hemoglobin A1C, fasting or postprandial glucose (Lovejoy et al 2002, Scott et al 2003, Tapsell et al 2004). However, a cross sectional study from Jiang et al. suggests that tree nut, peanut, and seed consumption is inversely associated with inflammatory biomarkers like CRP, IL-6 and fibrinogen, which may mediate cardiovascular disease and DM (Jiang et al 2006).

Several factors might be responsible for an apparent lack of effect of nuts on DM. We did not have data on the type of nuts or seeds consumed. Each type of nut, such as almonds, walnuts, hazelnuts, cashews, pistachios etc. has a unique nutrient profile, and possibly slightly different health benefit. For example, it has been shown that almonds reduce glycemic index of a white bread meal (Josse et al 2007) while walnuts consumption has been shown to increase the total plasma antioxidant capacity and decrease susceptibility to lipid peroxidation (Torabian et al 2009). Furthermore, we did not have data on the forms of nut consumed, i.e. salted, unsalted, honey coated or chocolate coated nuts. We did not have information to distinguish peanuts from tree nuts. These factors might attenuate the true association between nuts and DM. The fact that study participants were physicians, who may have different lifestyle habits than the general population limits us from generalizing these results to other populations. It is worthwhile to note that our study focused on nuts in general and not on legumes like peanuts in particular, which may be a substantial source of reported nut consumption by the study participants. Other limitations of the study include its inability to account for changes in frequency of nut consumption over time, possible over- or underreporting of nut consumption, and the lack of data on other foods and energy intake in this population to account for confounding by other dietary factors and energy intake. Thus, we cannot exclude the possibility of residual confounding in our data. In addition, we cannot exclude bias introduced by outcome misclassification as DM was ascertained through self-report. However, since participants were physicians, it is less likely that we have missed a substantial number of DM cases.

It is important to realize that nuts have been shown to have other health benefits. We have shown that nut consumption is associated with an 18% decreased risk of developing hypertension (Djousse et al 2009b). Similarly, a statistically significant decrease in total and LDL cholesterol has been shown in hyperlipidemic subjects consuming nuts (Mukuddem- Petersen et al 2005).Favorable effect of nuts on body weight (Bes-Rastrollo et al 2007), hypertension (Welty et al 2007) and sudden cardiac death (Albert et al 2002) has been demonstrated. Data from the Multi-Ethnic Study of Atherosclerosis suggests that nut consumption is associated with lower levels of C- reactive protein, interleukin-6 and fibrinogen, which may play a key role in cardiovascular disease and DM (Jiang et al 2006). Moreover, bias may account for the findings of our study; physicians at risk of developing diabetes, for example those with a family history of DM, or components of metabolic syndrome, might have higher nut consumption because of awareness of the above health benefits and thus leading the results towards the null. Despite these limitations, a large sample size, a 20 year follow up, standardized ascertainment and the fact that participants were physicians who could recognize early signs and symptoms of DM remain the strengths of our study.

In conclusion, our data do not provide evidence for an association between nut consumption and incident DM among US male physicians. Given multifactorial etiology of DM, any possible effect of nuts on DM cannot be ruled out in the general population.


We are indebted to the participants in the PHS for their outstanding commitment and cooperation and to the entire PHS staff for their expert and unfailing assistance.

Funding: The Physicians' Health Study is supported by grants CA-34944, CA-40360, CA-097193, HL-26490, and HL-34595, from the National Institute of Health, Bethesda, MD. No relationships with industry in respect to this project.


Conflict of interest: Drs Djoussé and Gaziano have received investigator-initiated grants from the National Institutes of Health. Dr. Gaziano has also received investigator-initiated grant from Veroscience. Dr. Kochar has no financial disclosures to make.


  • Albert CM, Gaziano JM, Willett WC, Manson JE. Nut consumption and decreased risk of sudden cardiac death in the Physicians' Health Study. Arch Intern Med. 2002;162:1382–1387. [PubMed]
  • 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. [PubMed]
  • Djousse L, Gaziano JM, Buring JE, Lee IM. Egg consumption and risk of type 2 diabetes in men and women. Diabetes Care. 2009a;32:295–300. [PMC free article] [PubMed]
  • Djousse L, Rudich T, Gaziano JM. Nut consumption and risk of hypertension in US male physicians. Clin Nutr. 2009b;28:10–14. [PMC free article] [PubMed]
  • Izzo R, de Simone G, Chinali M, Iaccarino G, Trimarco V, Rozza F, et al. Insufficient control of blood pressure and incident diabetes. Diabetes Care. 2009;32:845–850. [PMC free article] [PubMed]
  • Jenkins DJ, Kendall CW, Josse AR, Salvatore S, Brighenti F, Augustin LS, et al. Almonds decrease postprandial glycemia, insulinemia, and oxidative damage in healthy individuals. J Nutr. 2006;136:2987–2992. [PubMed]
  • Jiang R, Manson JE, Stampfer MJ, Liu S, Willett WC, Hu FB. Nut and peanut butter consumption and risk of type 2 diabetes in women. JAMA. 2002;288:2554–2560. [PubMed]
  • Jiang R, Jacobs DR, Jr., Mayer-Davis E, Szklo M, Herrington D, Jenny NS, et al. Nut and seed consumption and inflammatory markers in the multi-ethnic study of atherosclerosis. Am J Epidemiol. 2006;163:222–231. [PubMed]
  • Josse AR, Kendall CW, Augustin LS, Ellis PR, Jenkins DJ. Almonds and postprandial glycemia--a dose-response study. Metabolism. 2007;56:400–404. [PubMed]
  • Kochar J, Djousse L, Gaziano JM. Breakfast cereals and risk of type 2 diabetes in the Physicians' Health Study I. Obesity (Silver Spring) 2007;15:3039–3044. [PubMed]
  • Lovejoy JC, Most MM, Lefevre M, Greenway FL, Rood JC. Effect of diets enriched in almonds on insulin action and serum lipids in adults with normal glucose tolerance or type 2 diabetes. Am J Clin Nutr. 2002;76:1000–1006. [PubMed]
  • Mukuddem-Petersen J, Oosthuizen W, Jerling JC. A systematic review of the effects of nuts on blood lipid profiles in humans. J Nutr. 2005;135:2082–2089. [PubMed]
  • NIDDK National Diabetes Statistics, 2007 fact sheet. US Department of Health and Human Services, National Institutes of Health. 2008
  • Parillo M, Riccardi G. Diet composition and the risk of type 2 diabetes: epidemiological and clinical evidence. Br J Nutr. 2004;92:7–19. [PubMed]
  • Parker ED, Harnack LJ, Folsom AR. Nut consumption and risk of type 2 diabetes. JAMA. 2003;290:38–39. author reply 39–40. [PubMed]
  • 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;135:1114–1126. discussion 1127–1136. [PubMed]
  • Salvini S, Hunter DJ, Sampson L, Stampfer MJ, Colditz GA, Rosner B, et al. Food-based validation of a dietary questionnaire: the effects of week-to-week variation in food consumption. Int J Epidemiol. 1989;18:858–867. [PubMed]
  • Scott LW, Balasubramanyam A, Kimball KT, Aherns AK, Fordis CM, Jr., Ballantyne CM. Long-term, randomized clinical trial of two diets in the metabolic syndrome and type 2 diabetes. Diabetes Care. 2003;26:2481–2482. [PubMed]
  • Steering Committee of the Physicians' Health Study Research Group Final report on the aspirin component of the ongoing Physicians' Health Study. N Engl J Med. 1989;321:129–135. [PubMed]
  • Stein AD, Shea S, Basch CE, Contento IR, Zybert P. Consistency of the Willett semiquantitative food frequency questionnaire and 24-hour dietary recalls in estimating nutrient intakes of preschool children. Am J Epidemiol. 1992;135:667–677. [PubMed]
  • Tapsell LC, Gillen LJ, Patch CS, Batterham M, Owen A, Bare M, et al. Including walnuts in a low-fat/modified-fat diet improves HDL cholesterol-to-total cholesterol ratios in patients with type 2 diabetes. Diabetes Care. 2004;27:2777–2783. [PubMed]
  • Torabian S, Haddad E, Rajaram S, Banta J, Sabate J. Acute effect of nut consumption on plasma total polyphenols, antioxidant capacity and lipid peroxidation. J Hum Nutr Diet. 2009;22:64–71. [PubMed]
  • USDA Food availability spreadsheet-nuts. 2008. http://wwwersusdagov/data/foodconsumption/FoodAvailSpreadsheetshtm#nuts.
  • van Dam RM. The epidemiology of lifestyle and risk for type 2 diabetes. Eur J Epidemiol. 2003;18:1115–1125. [PubMed]
  • Welty FK, Lee KS, Lew NS, Zhou JR. Effect of soy nuts on blood pressure and lipid levels in hypertensive, prehypertensive, and normotensive postmenopausal women. Arch Intern Med. 2007;167:1060–1067. [PubMed]
  • Willett WC, Sampson L, Stampfer MJ, Rosner B, Bain C, Witschi J, et al. Reproducibility and validity of a semiquantitative food frequency questionnaire. Am J Epidemiol. 1985;122:51–65. [PubMed]