Long-Term Consumption of Nuts (Including Peanuts, Peanut Butter, Walnuts, and Other Nuts) in Relation to Risk of Frailty in Older Women: Evidence from a Cohort Study

Abstract

Background

Adherence to a healthy diet is inversely associated with frailty. However, the relationship between nuts, a key food group of Mediterranean diet, and frailty is unclear.

Objectives

This study aimed to evaluate the association between nut consumption and frailty in an aging female population.

Methods

This population-based observational study included nonfrail women (≥60 y old) in the NHS from 11 states of the United States. Outcome was incident frailty, defined as having ≥3 of the FRAIL components (fatigue, lower strength, reduced aerobic capacity, multiple chronic conditions, and significant weight loss) and assessed every 4 y from 1992 to 2016. From 1990 to 2014, FFQs were used to assess the intakes of peanuts, peanut butter, walnuts (added in 1998), and other nuts at 4-y intervals. Exposure was total nut consumption, calculated as the sum of intakes of peanuts, peanut butter, walnuts, and other nuts and categorized into <1 serving/mo, 1–3 servings/mo, 1 serving/wk, 2−4 servings/wk, and ≥5 servings/wk. The relations of intakes of peanuts, peanut butter, and walnuts with frailty were also investigated separately. Cox proportional hazards models were used to assess the associations between nut consumption and frailty after adjusting for age, smoking, BMI, EI, diet quality, and medication use.

Results

Among 71,704 participants, 14,195 incident frailty cases occurred over 1,165,290 person-years. The adjusted HR (95% CI) for consuming ≥5 servings/wk of nuts was 0.80 (0.73, 0.87), as compared with <1 serving/mo. Higher intakes of peanuts and walnuts, but not peanut butter, were also inversely associated with frailty.

Conclusions

This large prospective cohort study showed a strong and consistent inverse association between regular nut consumption and incident frailty. This suggests that nut consumption should be further tested as a convenient public health intervention for the preservation of health and well-being in older adults.

Introduction

Frailty, defined as an increasing vulnerability to stressful events because of the depletion in functional reserves [1], is a hallmark event of aging. In high-income countries, the prevalence of frailty is ∼11% among community-dwelling older adults [2]. Subpopulations that are disproportionately affected by frailty include women [3] and long-term care residents [4]. Frailty independently predicts many adverse outcomes, including mortality, disability, falls, and poor surgical outcomes, and is related to increased healthcare utilization and costs [1, [5], [6], [7], [8], [9], [10]]. Frailty also has detrimental effects on the physical and social functioning of older adults and diminishes the quality of life [6, 11].

Emerging evidence suggests that adherence to high-quality dietary patterns such as Mediterranean diet is inversely associated with risk of frailty [[12], [13], [14], [15], [16]]. Nuts are a main food component of the Mediterranean diet and contain a wide array of nutrients such as polyunsaturated fats, vitamins, and anti-inflammatory phytochemicals that are involved in maintaining key physiological functions to prevent the development of frailty [[17], [18], [19], [20], [21], [22]]. However, no study has specifically investigated the relationship between amount of nut consumption and frailty risk. In this study, we aimed to evaluate the association between nut consumption (including peanuts, peanut butter, walnuts, and other nuts) and frailty in a large population of older women from the Nurses' Health Study (NHS). We hypothesized that higher nut intake would be associated with lower risk of frailty, independent of other risk factors for aging.

Methods

Study cohort

The NHS is an ongoing prospective cohort study of chronic diseases in women [23, 24]. It was established in 1976 and enrolled 121,700 nurses aged 30–55 y (98% White), who resided in New York, California, Pennsylvania, Ohio, Massachusetts, New Jersey, Michigan, Texas, Florida, Connecticut, or Maryland at the time of enrollment. Participants completed a baseline questionnaire that collected data on demographics, lifestyle factors, and medical history. Every 2 y, they receive a follow-up questionnaire to collect updated information on diseases and health-related variables, including weight, new disease diagnoses, and health status.

The study protocol was approved by the institutional review boards of the Brigham and Women’s Hospital and the Harvard T.H. Chan School of Public Health.

Dietary assessment

FFQ, administered at 4-y intervals from 1990 to 2014, were used to assess the frequency of consuming peanuts, peanut butter, walnuts (added to the questionnaire in 1998), and other nuts during the past year (in a serving unit of 1 ounce for peanuts/walnuts/other nuts or 2 tablespoons for peanut butter). FFQ is one of the most common instruments for dietary assessment and has been well-validated in a series of studies conducted in the NHS [25, 26].

The primary exposure of interest is total amounts of nut consumption. Total nut intake at each study cycle was computed as the sum of intakes of peanuts, peanut butter, walnuts, and other nuts. To obtain a measure of long-term nut intake, we averaged the total nut intake from all available FFQs before each frailty assessment. Because peanut butter can often be highly processed, we also considered an alternative measure of total nut intake after excluding peanut butter. Total nut intake was categorized into <1 serving/mo (reference group), 1–3 servings/mo, 1 serving/wk, 2–4 servings/wk, and ≥5 servings/wk [[27], [28], [29]]. Further, we explored the intakes of peanuts, peanut butter, and walnuts in separate analyses. In these analyses, the 2–4 servings/wk and ≥5 servings/wk categories of nut intake were combined because of few frailty cases in these categories.

Frailty assessment

Frailty was assessed using the questionnaire-based FRAIL scale, a well-validated instrument for frailty assessment [30]. The 5 components of FRAIL scale included fatigue, lower strength, reduced aerobic capacity, multiple chronic conditions, and significant weight loss during the previous year. Data on the components of the FRAIL scale were available every 4 y from 1992 to 2016. Incident frailty was defined as having 3 or more of the FRAIL components. The definitions of each of the 5 frailty components were: 1) Fatigue: from 1992 to 2000, defined as answering “some of the time” or “none of the time” to the question “Did you have a lot of energy over the past year?”; in 2004, defined as answering “moderate amount” or “all of the time” to the question “I could not get going”; from 2008 to 2016, defined as answering “no” to the statement “Do you feel full of energy?”; 2) Lower strength: defined as answering “yes, limited a lot” to the question “In a normal day, is your health a limitation to walk up 1 flight of stairs?”; 3) Reduced aerobic capacity: defined as answering “yes, limited a lot” to the question “In a normal day, is your health a limitation to walk several blocks or several miles?”; 4) Chronic conditions: defined as having reported diagnosis of 5 or more of the following conditions: cancer, hypertension, type 2 diabetes, angina, MI, stroke, congestive heart failure, peripheral artery disease, asthma, chronic obstructive lung disease, arthritis, Parkinson disease, kidney disease, or depression; 5) Significant weight loss: defined as ≥5% decrease in weight reported in 2 consecutive follow-up cycles.

Covariate assessment

Data on the date of birth and height were collected using the NHS baseline questionnaire. Data on weight and smoking status were collected in the baseline questionnaire and updated at each 2-y study cycle. BMI was calculated as kg/m². PAL were assessed using self-reported estimates of time spent per week on a list of leisure-time activities, such as brisk walking, running, and bicycling, from which the weekly expenditure of metabolic equivalents was derived. Alternate Healthy Eating Index (AHEI-2010) excluding the nut component, a measure of overall diet quality [31], and total EI were estimated using the FFQs administered at the same study cycles that the nut consumption data were available.

Population for analysis

We included all NHS participants aged 60 y and older who were not frail in 1992 (baseline). Women younger than 60 y old at baseline entered the study when they turned 60 y old during the follow-up. We excluded women with unreasonably high (>3500 kcal/d) or low (<500 kcal/d) caloric intake or missing data on amount of nut intake.

Statistical analysis

We used Cox proportional hazards models to evaluate the associations between the amounts of nut consumption and incident frailty. Participants contributed person-time from the first return of a questionnaire for frailty assessment at or after the age of 60 y through the occurrence of incident frailty, death, the end of study period (2018) or loss to follow-up, whichever came first. We used age and study cycle (as a proxy for calendar periods) as the time scale to allow for the best control of confounding by age and calendar time. Counting process data structure was used to handle the left truncation issue and time-varying covariates. Nut intake was modeled categorically to determine if there was a nonlinear relationship between nut consumption and risk of frailty. Furthermore, we modeled nut intake as a continuous variable (created using the median value of each nut consumption category) to test for linear trend between nut consumption and frailty risk. The proportional hazards assumption was assessed using interaction terms between age indicator variables (in 5- or 10-y age brackets) and amount of nut intake. Four sets of models were used to control for confounding: 1) The base model adjusted for age and calendar periods; 2) The second model additionally adjusted for time-updated smoking, BMI, and EI; 3) The third model additionally adjusted for AHEI-2010 score; and 4) The last model additionally adjusted for medication use, including postmenopausal hormone replacement therapy, aspirin, diuretics, β-blockers, calcium channel blockers, angiotensin-converting enzyme inhibitors, other antihypertensive medication, statins and other cholesterol-lowering drugs, insulin, and oral hypoglycemic medication (yes/no), to address the fact that individuals with chronic conditions may be at higher risk of developing frailty. To best account for potential nonlinearity in the relationship between confounders and frailty risk, we included indicators of quintiles of BMI, EI, and AHEI-2010 score. Because PAL is essentially a component of frailty, we additionally adjusted for this variable in sensitivity analyses to understand its impact on the relation of diet to frailty. Data were carried forward for ≤1 study cycle when a question for a desired variable was not asked in the current cycle. A missing indicator was then generated and included in the analytical model as there were few missing data. We also conducted a sensitivity analysis using multiple imputations to assess the impact of missing data on the association between total nut consumption and incident frailty. Five imputed datasets were obtained using the Markov chain Monte Carlo method. The imputation model includes frailty, amount of nut consumption, and all covariates considered in the fully adjusted Cox regression model.

To understand whether relations of nut consumption and frailty differ for nonfrail and prefrail individuals, we repeated the analysis in participants who had none of the 5 components of FRAIL scale at baseline. We assessed the association between nut consumption and each of the 5 components of frailty to provide further insights into relations. We also conducted analyses in which we lagged nut consumption (by 6 y) to reduce possible reverse causation because of change in the amount of nut consumption induced by early signs of frailty. Lastly, to assess the impact of tooth loss on our findings, we performed analyses restricting to participants without severe tooth loss, defined as having ≥10 original teeth (assessed in 1992, 2012, and 2016).

All analyses were conducted using SAS software, Version 9.4 (SAS Institute Inc) with 2-sided tests at the significance level of 0.05.

Results

Participant characteristics

A total of 71,704 participants were included in the analysis for the association between total nut consumption and frailty (Figure 1). Of these, 71,685 and 69,796 participants were included in the analyses of peanuts/peanut butter and walnuts consumption, respectively (Figure 1).

FIGURE 1.

Selection of eligible NHS participants.

Table 1 summarizes the age-standardized participant characteristics at baseline (1992) by the amount of total nut consumption (including peanut butter). This table is restricted to 27,606 participants who were ≥60 y old in 1992. Age-standardized characteristics for the entire cohort are shown in Supplemental Table 1. As compared with individuals with lower total nut intake, individuals who ate more nuts had lower mean BMI, higher PALs, higher EI, better diet quality (as indicated by higher mean AHEI-2010 score) and were less likely to be current smokers. Approximately 30% of the study population were prefrail (having ≥1 FRAIL components) at baseline and this did not differ meaningfully across the amounts of total nut consumption.

TABLE 1.

Age-standardized characteristics of NHS participants who were ≥60 y old at baseline (1992)

	Total nut consumption (including peanut butter) (n = 27,606)
	<1 serving/mo (n = 8436)	1–3 servings/mo (n = 5324)	1 serving/wk (n = 9607)	2–4 servings/wk (n = 2837)	≥5 servings/wk (n = 1402)
Age, y, mean (SD)1	64.1 (2.63)	64.1 (2.65)	64.1 (2.64)	64.2 (2.61)	64.1 (2.63)
BMI, kg/m2, mean (SD)	25.8 (4.67)	25.7 (4.59)	25.6 (4.5)	25.2 (4.39)	24.9 (4.40)
Current smoker, %	15	14	14	12	14
EI, kcal/d, mean (SD)	1543 (465)	1643 (461)	1804 (481)	1966 (504)	2137 (537)
PAL, MET-h/wk, median (Q1, Q3)	9.8 (3.2, 22.4)	10.5 (3.9, 23.4)	11.6 (4.2, 25.3)	13.8 (4.9, 27.9)	13.4 (4.9, 28.5)
AHEI-2010 score, mean (SD)	51.1 (10.7)	51.0 (10.3)	50.8 (10.2)	51.0 (9.9)	52.6 (10.5)
Current aspirin use, %	46	49	50	50	50
FRAIL components, %
None	72	74	76	77	75
1 (prefrail)	21	21	19	18	20
2 (prefrail)	7	5	5	5	5

All values are standardized to the age distribution of the study population unless otherwise noted. AHEI-2010, Alternative Healthy Eating Index 2010 excluding the nut component; FRAIL, fatigue, lower strength, reduced aerobic capacity, multiple chronic conditions and significant weight loss; MET, metabolic equivalents; Q1, 25^th quartile; Q3, 75^th quartile.

¹Value is not age-adjusted.

The association between total nut consumption and frailty

A total of 14,195 incident cases of frailty developed over a median follow-up of 16 y (1,165,290 person-years). The adjusted HRs of total amount of nut consumption (including peanut butter) and risk of frailty are shown in Table 2. Frequent nut consumption was strongly associated with lower risk of frailty with clear linear trends. After adjusting for age, BMI, EI, and smoking, individuals who consumed ≥5 servings/wk of nuts had an HR (95% CI) of 0.70 (0.64, 0.76), as compared with the <1 serving/mo group. Additional adjustment for AHEI-2010 attenuated this association somewhat, but findings remained statistically significant [HR (95% CI): 0.80 (0.74, 0.88)]. Further adjustment for medication use did not alter this association. Similar findings were observed after excluding peanut butter from the calculation of total nut intake (Table 2).

TABLE 2.

Adjusted associations of the total nut consumption with incident frailty in the NHS participants

		Amount of consumption (HRs1 [95% CI])
	Models	<1 serving/mo	1–3 servings/mo	1 serving/wk	2–4 servings/wk	≥5 servings/wk	P-trend
Total nut consumption (including peanut butter)2	Incident frailty, n	2675	2616	6071	2099	734	—
	Person-years	238,736	217,317	485,416	157,766	66,055	—
	Age-adjusted	Ref	0.94 (0.89, 1.00)	0.87 (0.83, 0.91)	0.80 (0.76, 0.85)	0.65 (0.60, 0.71)	<0.001
	+Smoking, BMI, EI	Ref	0.93 (0.89, 0.99)	0.87 (0.83, 0.91)	0.82 (0.78, 0.88)	0.70 (0.64, 0.76)	<0.001
	+AHEI-2010	Ref	0.95 (0.90, 1.00)	0.90 (0.86, 0.95)	0.89 (0.84, 0.95)	0.80 (0.74, 0.88)	<0.001
	+Medication use	Ref	0.95 (0.90, 1.00)	0.89 (0.85, 0.94)	0.88 (0.83, 0.94)	0.80 (0.73, 0.87)	<0.001
Total nut consumption (excluding peanut butter)3	Incident frailty, n	5893	2904	4088	975	335	—
	Person-years	470,506	247,179	330,723	83,471	33,410	—
	Age-adjusted	Ref	0.88 (0.84, 0.92)	0.80 (0.77, 0.83)	0.68 (0.63, 0.72)	0.62 (0.55, 0.69)	<0.001
	+Smoking, BMI, EI	Ref	0.89 (0.85, 0.93)	0.82 (0.79, 0.86)	0.73 (0.68, 0.78)	0.70 (0.63, 0.79)	<0.001
	+AHEI-2010	Ref	0.92 (0.88, 0.96)	0.88 (0.85, 0.92)	0.83 (0.77, 0.89)	0.84 (0.75, 0.94)	<0.001
	+Medication use	Ref	0.91 (0.87, 0.95)	0.87 (0.84, 0.91)	0.83 (0.77, 0.89)	0.84 (0.75, 0.94)	<0.001

AHEI-2010, Alternative Healthy Eating Index 2010 excluding the nut component; Ref, reference group.

¹Reference category for HR estimation is the <1 serving/mo category.

²Total nut consumption (including peanut butter) was calculated as the sums of intake of peanuts, peanut butter, walnuts, and other nuts.

³Total nut consumption (excluding peanut butter) was calculated as the sums of intake of peanuts, walnuts, and other nuts.

The association of peanuts, peanut butter, and walnuts consumption with frailty

Table 3 shows the associations of peanuts, peanut butter, and walnuts intake with frailty. Eating more peanuts and walnuts was associated with a substantial decrease in risk of frailty. Compared to <1 serving/mo, the adjusted HRs (95% CIs) for consuming ≥2 servings/wk of peanuts and walnuts were 0.84 (0.74, 0.94) and 0.85 (0.75, 0.97), respectively. By contrast, higher amounts of peanut butter consumption were not associated with reduced risk of frailty.

TABLE 3.

Adjusted associations of peanut, peanut butter, and walnut consumption with incident frailty in the NHS participants

		Amount of consumption (HRs1 [95% CI])
	Models	<1 serving/mo	1–3 servings/mo	1 serving/wk	≥2 servings/wk	P-trend
Peanut	Incident frailty, n	9801	2209	1874	308	—
	Person-years	759,931	217,235	151,472	35,539	—
	Age-adjusted	Ref	0.86 (0.82, 0.90)	0.84 (0.80, 0.89)	0.73 (0.65, 0.82)	<0.001
	+Smoking, BMI, EI	Ref	0.86 (0.82, 0.91)	0.88 (0.83, 0.92)	0.80 (0.71, 0.89)	<0.001
	+AHEI-2010	Ref	0.88 (0.84, 0.92)	0.91 (0.86, 0.95)	0.84 (0.75, 0.94)	<0.001
	+Medication use	Ref	0.87 (0.83, 0.91)	0.90 (0.86, 0.95)	0.84 (0.74, 0.94)	<0.001
Peanut butter	Incident frailty, n	6907	3194	3637	454	—
	Person-years	610,856	244,499	270,918	37,689	—
	Age-adjusted	Ref	0.99 (0.95, 1.04)	1.00 (0.96, 1.04)	0.94 (0.85, 1.03)	0.42
	+Smoking, BMI, EI	Ref	0.98 (0.94, 1.02)	0.99 (0.95, 1.04)	0.95 (0.86, 1.04)	0.36
	+AHEI-2010	Ref	0.98 (0.94, 1.02)	0.99 (0.95, 1.03)	0.94 (0.86, 1.04)	0.32
	+Medication use	Ref	0.97 (0.93, 1.01)	0.98 (0.94, 1.03)	0.94 (0.85, 1.03)	0.32
Walnut	Incident frailty, n	9790	1508	1333	235	—
	Person-years	700,957	149,429	94,014	23,022	—
	Age-adjusted	Ref	0.82 (0.77, 0.86)	0.76 (0.72, 0.81)	0.63 (0.55, 0.71)	<0.001
	+Smoking, BMI, EI	Ref	0.84 (0.80, 0.89)	0.80 (0.75, 0.85)	0.70 (0.61, 0.79)	<0.001
	+AHEI-2010	Ref	0.90 (0.85, 0.95)	0.91 (0.85, 0.96)	0.85 (0.74, 0.97)	<0.001
	+Medication use	Ref	0.89 (0.84, 0.94)	0.90 (0.85, 0.96)	0.85 (0.75, 0.97)	<0.001

AHEI-2010, Alternative Healthy Eating Index 2010 excluding the nut component; Ref.

¹Reference category for HR estimation is the <1 serving/mo category.

Sensitivity analyses

Our findings of inverse associations between the amount of nut consumption and risk of frailty were consistent across multiple sensitivity analyses. First, higher amounts of nut consumption remained associated with lower HRs of frailty in the 6-y lagged analyses (Supplemental Table 2). Of the 5 frailty components, an inverse association with higher amount of nut consumption was noted for lower strength, decline in aerobic capacity, and having multiple chronic conditions (Supplemental Tables 3–7). Additional adjustment for PALs slightly attenuated the association between higher nut intake and frailty, but this did not materially change the study results (Supplemental Table 8). Restricting the analytical cohorts to robust individuals with none of the FRAIL components (n = 55,977) (Supplemental Table 9) and to individuals without severe tooth loss (n = 59,662) (Supplemental Table 10) did not result in material changes to the study findings. Lastly, using multiple imputation yielded an adjusted HR (95% CI) of 0.83 (0.76, 0.91) for consuming ≥5 servings/wk of nuts, after adjusting for smoking, BMI, EI, and AHEI-2010. This is comparable to the results reported in the primary analysis.

Discussion

In this large-scale, prospective cohort study, we found that nut consumption was strongly associated with a lower risk of frailty among older women, after adjusting for health and lifestyle factors. Specifically, individuals who consumed ≥5 servings/wk of nuts (including peanuts, peanut butter, walnuts, and other nuts) had ∼20% lower risk of developing frailty as compared with those who consumed <1 serving/mo. In addition, there was a linear trend, such that each increasing quantity of nuts consumed was related to an incremental reduction in frailty risk. An inverse association with frailty risk was also found for peanuts and walnuts, but not for peanut butter. Nuts are a relatively simple health intervention, thus our findings suggest that further testing of nuts to prevent frailty, a common and debilitating condition in older populations, could have important public health implications in our rapidly aging society.

There is growing interest in understanding the role of diet in the etiology of frailty and other aging-related outcomes. Many studies consistently showed that higher adherence to Mediterranean dietary pattern that is rich in intake of fruits, vegetables, nuts, grains, and seafood was associated with reduced risks of frailty [15, 16, [32], [33], [34], [35], [36]]. However, few studies investigated the relationship between individual food groups and the incidence of frailty. Nuts are a key component of a Mediterranean diet. Emerging evidence suggests health benefits of regular nut consumption in delaying the decline of functional outcomes and promoting health and well-being in aging. In recent prospective cohort studies, higher nut intake was inversely associated with risk of impaired mobility [37] and positively associated with healthy aging, a composite outcome defined as having no history of chronic diseases, no reported memory impairment, no physical disabilities, and intact mental health [38]. Eating more nuts, evaluated as part of healthy diet patterns, was also associated with higher percent of skeletal muscle mass [39], although this association was not observed in studies of adverse functional outcomes such as physical impairment, falls or frailty [33, 39, 40]. However, with short follow-up periods in these studies, it may have been difficult to identify relations with chronic conditions that typically take a long time to manifest, such as frailty.

Several mechanisms may explain the observed inverse association between nut consumption and frailty. First, nuts are rich in essential nutrients, including plant-based proteins, mono- and poly-unsaturated fats, vitamins, and minerals. Current evidence suggests that these nutrients may be involved in protein synthesis and prevention of protein breakdown in muscles, thus delaying the onset of sarcopenia, a contributing physical component to frailty [1, 41, 42]. Second, the rich content of unsaturated fats in nuts can induce favorable lipid profiles that have known cardiometabolic benefits, which help to preserve aerobic capacity [[43], [44], [45], [46], [47]]. In addition, nuts are energy dense. Higher intake of nuts may, therefore, prevent unintentional weight loss as a consequence of undernutrition, a common condition in older adults [48]. Nuts also stimulate satiety after ingestion [49], which can reduce risk of obesity, an important predictor of frailty especially when it occurs concurrently with reduced muscle mass and strength [50, 51].

We also found that both peanuts and walnuts were associated with lower risks of frailty. Walnuts contain high content of phenolic compounds with antioxidative and anti-inflammatory properties and are the only tree nut with rich source of alpha-linolenic acid (ALA), an essential omega-3 FA [52]. ALA has potent anti-inflammatory properties and reduces both acute and chronic inflammation in animal models [[53], [54], [55]]. In human feeding studies, supplementation of high-fat meals with walnuts resulted in acute postprandial and long-term suppression of proinflammatory biomarkers that are implicated in the pathogenesis of frailty, such as intercellular adhesion molecule 1 (ICAM-1) and IL-6 [[56], [57], [58], [59]]. Given the detrimental effect of chronic inflammation in the aging process, regular walnut consumption may reduce risk of frailty through suppressing the heightened immune state that contributes to the development of frailty. Unlike peanuts and walnuts, we found that peanut butter was not related to frailty risk. Many different types of peanut butter can be highly processed and included added sugar or added fats. These may potentially influence apparent benefits of the nuts.

Our study has limitations. First, the amounts of nut consumption were estimated from self-administered FFQs [25]. There may be random measurement error in the self-reports, which could result in bias to the null in our observed nut-frailty association; thus, our findings may under-estimate the relation of nut intake to frailty. The second limitation is the use of the FRAIL frailty scale, which is also based on self-reported components [30]. However, validation studies have shown that the FRAIL scale had good predictive accuracy for various disease and functional outcomes [30, 60, 61]. Residual and unmeasured confounding is also possible given the observational nature of our study. However, we have considered a comprehensive list of key confounders using highly flexible functional forms in statistical modeling. Nonetheless, it is not possible to eliminate confounding here, and our results should be interpreted cautiously. Our study is also not powered to study the association between categories of nut consumption higher than ≥5 servings/wk and frailty. Lastly, the NHS enrolled highly educated, mostly White, female nurses. The lack of diversity in the study population may limit the generalizability of the study findings. However, the NHS has been utilized for substantial previous research, with results that have been consistent with the broader literature in other racial and ethnic groups.

In conclusion, in this large prospective cohort of women of age ≥60 y, we found strong and consistent associations between frequent nut consumption and lower risk of frailty. Our study adds to the broad literature supporting regular nut consumption as a possible public health intervention in preserving health and well-being in aging populations.

Author disclosures

An unrestricted research award from the California Walnut Commission, Sacramento, CA, US, was provided to FG. None of the other authors have a commercial or other association that might pose a conflict of interest.

Funding

This work was supported by grant UM1 CA186107 from the NIH. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Data Availability

Data described in the manuscript, code book, and analytic code will be made available upon request pending application and approval.

Appendix A. Supplementary data

The following is the Supplementary data to this article: