Adherence to 2020 to 2025 Dietary Guidelines for Americans and the Risk of New-Onset Female Gout

Key Points

Question  Is adherence to guideline-based healthy eating patterns associated with the risk of incident female gout?

Findings  In this prospective cohort study of 80 039 women from the Nurses’ Health Study followed up to 34 years, the top-level adherence to multiple healthy eating patterns was associated with 12% to 32% lower risk of incident gout; the corresponding risk reduction reached 65% to 68% when combined with normal weight or not using diuretics.

Meaning  These findings extend the pleotropic benefits of the 2020 to 2025 Dietary Guidelines for Americans to female gout prevention, with multiple healthy diets that can be adapted to individual food traditions, preferences, and comorbidities.

Importance  Female-specific gout data are scarce despite perceived differences from males in its risk factors and disproportionate worsening in disease and comorbidity burden globally. The 2020 to 2025 Dietary Guidelines for Americans recommend multiple healthy eating patterns for prevention of cardiovascular-metabolic outcomes, which may also be relevant to the prevention of female gout.

Objective  To examine the associations of dietary scores for the latest guideline-based healthy eating patterns with risk of incident female gout.

Design, Setting, and Participants  This prospective cohort study included 80 039 US women in the Nurses’ Health Study followed up through questionnaires every 2 years starting from 1984. Participants had no history of gout at baseline, and the study used questionnaire responses through 2018. Statistical analyses were performed over September 2020 to August 2021.

Exposures  Four healthy eating patterns: Dietary Approaches to Stop Hypertension (DASH), Alternate Mediterranean Diet Score, Alternative Healthy Eating Index (AHEI), and Prudent, plus Western (unhealthy) for comparison, with scores derived from validated food frequency questionnaires.

Main Outcomes and Measures  Incident, physician-diagnosed female-specific gout.

Results  During 34 years of follow-up, we documented 3890 cases of incident female gout. Compared with the least-adherent quintile, women most adherent to healthy diets had significantly lower risk of incident gout, with multivariable-adjusted hazard ratios (HRs) 0.68 (95% CI, 0.61-0.76) (DASH), 0.88 (95% CI, 0.80-0.98) (Mediterranean), 0.79 (95% CI, 0.71-0.89) (AHEI), and 0.75 (95% CI, 0.73-0.90) (Prudent); all P for trend ≤.009. Conversely, women with highest-quintile Western diet score had 49% higher risk of gout (HR, 1.49; 95% CI, 1.33-1.68], P <.001). When combined, the most DASH-diet adherent women with normal body mass index (BMI; calculated as weight in kilograms divided by height in meters squared) had a 68% lower risk of gout compared with the least adherent women with overweight or obese BMI; the corresponding risk reduction was 65% combining high DASH diet adherence with no diuretic use.

Conclusions and Relevance  These large-scale, long-term prospective cohort findings extend the pleotropic benefits of the 2020 to 2025 Dietary Guidelines for Americans to female gout prevention, with multiple healthy diets that can be adapted to individual food traditions, preferences, and comorbidities.

Gout, a metabolic condition causing the most common form of inflammatory arthritis, leads to excruciatingly painful recurrent flares and joint damage.^1,2 Described by Hippocrates during the Golden Age of Greece, gout had been considered a disease of the affluent, middle-aged men of the upper class (Patricain malady, disease of kings) who could afford the comforts—and excesses—of life.³ However, with Western lifestyles growing increasingly prevalent over recent decades, the frequency of gout has increased substantially.^4-6 Indeed, recent analyses of the Global Burden of Disease Study noted a disproportionate increase in gout burden among women between 1990 and 2017,⁷ and highlighted the need for intensive dietary management and other preventive strategies to mitigate these trends.^7,8 Furthermore, given that female gout is associated with key comorbidities more frequently than male gout, including coronary artery disease, hypertension, type 2 diabetes, and chronic kidney disease,^9-14 there is a pressing need for pleiotropic approaches that can address both gout and these comorbidities, conditions historically overlooked and undertreated in women.^15,16

To this end, adoption of cardiometabolically healthy lifestyles represents an attractive option to simultaneously address gout and cardiovascular risk. In particular, the healthy eating patterns described in the 2020 to 2025 Dietary Guidelines for Americans¹⁷ and recommended by professional societies,¹⁸ including the Dietary Approaches to Stop Hypertension (DASH) and Mediterranean, are intended to accommodate a variety of personal and cultural food preferences, and have been shown to improve cardiovascular risk factors^19-21 and prevent adverse cardiovascular-metabolic outcomes and malignant diseases.^17,20,22-24 Isocaloric forms of these diets also have the capacity to lower serum urate and modulate inflammation,^25-27 whereas calorie-restricted forms may further reduce serum urate by lowering insulin resistance or adiposity.²⁸ For example, several analyses of dietary randomized interventional trials have found that healthy diets can lower serum urate levels, with larger effects among those with hyperuricemia, thus at risk of developing gout.^25,29 Furthermore, a prospective cohort study of US men found the DASH diet was associated with a lower risk of incident gout (adjusting for body mass index [BMI, calculated as weight in kilograms divided by height in meters squared]) and total energy intake), and the Western diet with a higher risk.³⁰ However, no relevant data are available among women, despite the perceived differences from men in risk factors (eg, higher frequency of diuretic use^10-13 and more frequent obesity among women,^11-13 but lower consumption of alcohol^11-13) and cross-sectional data suggesting weaker or null associations between diet and serum urate among women.^31,32 Our objective was to investigate adherence to 4 healthy eating patterns, which reflect the 2020 to 2025 Dietary Guidelines for Americans, and risk of incident female gout among 80 039 at-risk women followed for more than 2.3 million person-years.

The Nurses’ Health Study (NHS) is an ongoing longitudinal cohort study that began in 1976 when 121 700 female registered nurses age 30 to 55 completed mailed questionnaires about their medical history, lifestyle, and other risk factors. Thereafter, participants have regularly completed follow-up questionnaires to update lifestyle exposures and identify newly diagnosed diseases; response rate has exceeded 90% of eligible person time. Starting in 1984, participants were also asked to complete a validated food frequency questionnaire (FFQ) every 4 years with detailed questions about average intake of individual foods and beverages during the previous year.^33,34 Completion of the self-administered questionnaires was considered to imply informed consent. The current analysis was approved by the Massachusetts General Brigham institutional review board.

Using data collected in the FFQ, we derived scores for 4 healthy dietary patterns—DASH, Mediterranean, Alternative Healthy Eating Index (AHEI), and Prudent, along with the Western dietary pattern, a contrasting unhealthy exposure. These scores have been used extensively in the NHS^20,22,35; further details on their components and derivation of scores are in the eMethods in the Supplement. Briefly, DASH score was based on participants’ reported intake of items emphasized or minimized in the DASH diet, including fruits, vegetables, low-fat dairy, and sugar-sweetened beverages.²² The Mediterranean assigns positive scores for vegetables, fruits, whole grains, nuts, legumes, and fish, and reverse scores for red and processed meat.^20,35 The AHEI 2010 assigns points to foods and nutrients that have consistently been associated with lower risk of chronic disease in clinical and epidemiologic studies.³⁶ Prudent and Western diet scores were derived using a principal component analysis that grouped items from the FFQ into approximately 40 groups.³⁷ Prudent pattern was characterized by higher intake of fruits, vegetables, whole grains, poultry, and low-fat dairy, and Western pattern by higher intake of red and processed meats, refined grains, sweets, desserts, and high-fat dairy.

Along with the FFQ, participants completed biennial questionnaires inquiring about weight, menopausal status, tobacco use, regular use of medications (including diuretics [thiazide or loop] and hormone replacement therapy [HRT]) and medical comorbidities (including hypertension).²⁰ Participant BMI was calculated by dividing the most recently updated weight in kilograms by the square of the baseline height in meters. The medical and lifestyle information provided by the participants through these questionnaires have been shown reliable in validation studies and able to predict relevant future diseases, such as cardiovascular disease,^20,22 type 2 diabetes,³⁸ and gout.³⁹

Incident gout was based on the nurses’ report of new-onset, physician-diagnosed gout. In 1982, 1984, 1986, 1988, 2002, and thereafter, biennial questionnaires asked the participants whether they had received a diagnosis of gout from a physician, and if so, the date of first occurrence.^39,40

We computed person-time in years for each participant from the return date of the 1984 questionnaire to date of gout diagnosis, death, or end of follow-up period (June 2018), whichever came first. To better represent the average dietary intake by individual participants over the long term, we used the cumulative average intake based on the dietary information from baseline (1984 questionnaire, the first with detailed dietary intake data) to the latest point of follow-up as a time-varying variable.^39,41,42 We used Cox proportional hazards modeling to estimate the risk for incident gout, dividing the dietary pattern scores into quintiles, with the highest quintile indicating greatest adherence to a given pattern, and, as done previously, also assessing scores continuously (per 25-percentile increment,³⁵ and per standard deviation [SD],⁴³ to permit comparison across patterns). Multivariable models were adjusted in a time-varying manner for the following: age, menopause, use of hormone therapy, history of hypertension, BMI, diuretic use, alcohol intake, coffee intake, and total energy (calorie) intake.⁴⁴ We also conducted stratified analyses to evaluate for potential effect modification according to diuretic use, BMI (<25 or ≥25), alcohol consumption (yes or no), menopausal status, and postmenopausal HRT use (ever, yes or no). These variables were selected based on documented higher frequencies of diuretic use and obesity, and lower frequency of alcohol use, among women with gout compared with men,^10-13 and increased (decreased) risk of gout associated with menopause and HRT,⁴⁰ respectively. We compared frequencies among women in the current analysis with men with incident gout (n = 1741) in our prior prospective analysis (ie, 20% with diuretic use, 20% obese, and 81% with alcohol use).⁴⁵ The significance of the interaction was tested using the likelihood ratio test by comparing a model with the main effects of each dietary pattern and the stratifying variable and the interaction terms, with a reduced model with the main effects only. These factors (diuretic use, BMI, and alcohol use) were also assessed for the joint (ie, combined) associations with each dietary pattern. All statistical analyses were conducted in SAS statistical software (version 9.4, SAS Institute; Inc). All P values are 2-sided.

Over 34 years of follow-up, we documented 3890 cases of incident gout among 80 039 women (mean age at baseline, 50.5 years; mean BMI, 25.0 m/kg2). Table 1 and eTable 1 in the Supplement show baseline characteristics of the cohort according to quintiles of dietary pattern scores. For each of the 4 healthy diets, women in the highest quintile tended to be older than those in the lowest quintile; the reverse was seen for the Western diet. Women in the highest quintile of DASH and AHEI scores tended to have lower intakes of alcohol and coffee compared with the lowest quintile, whereas for the Mediterranean diet, there was little difference in alcohol and coffee intake across quintiles. Total energy intake was highest for women in the highest quintile of the Western score, followed by the highest quintile of the Prudent and Mediterranean dietary scores. Spearman correlation coefficients between the 4 healthy diet scores ranged from 0.47 to 0.68 (all P < .001) (eTable 2 in the Supplement). The nondietary variables were well-balanced between quintiles within each dietary pattern score with exception of HRT use, which tended to be higher in quintile 5 of the healthy diets and quintile 1 of the Western diet (Table 1).

The 4 healthy dietary patterns (DASH, Mediterranean, AHEI, and Prudent) were consistently associated with a lower risk of incident gout, as summarized in Table 2. Numerically, the risk reduction was greatest with the DASH dietary pattern, with a 32% reduced risk of incident gout in the most adherent quintile compared with the least adherent (multivariable HR, 0.68; 95% CI, 0.61-0.76; P for trend <.001; eFigure 1 in the Supplement). The corresponding risk reduction from the 3 other dietary patterns for the highest compared with the lowest quintile ranged from 12% to 25% (Figure 1). Results were similar when using continuous scores (Table 2).

In contrast to the 4 healthy patterns, the highest quintile Western diet score (ie, most unhealthy) was associated with a 49% increased risk of incident gout (HR, 1.49; 95% CI, 1.33-1.68; P for trend <.001) compared with those with the lowest quintile (ie, healthiest) (Table 2).

At the time of gout diagnosis, 1659 women (43%) in our cohort were using diuretics, 1593 (41%) were obese, and 2210 (52%) were users of alcohol. The associations of each diet with the risk of incident gout persisted when stratifying women by these variables, except for Mediterranean diet among overweight or obese women, which appeared to have minimal or no effect on gout risk (Table 3). Additional subgroup analyses according to menopausal status (and HRT use) showed similar results (eTable 3 in the Supplement).

In the joint analysis, the most DASH-diet adherent women with normal BMI had a 68% lower risk of gout (HR, 0.32; 95% CI, 0.26-0.38) compared with the least adherent women with overweight or obese BMI (Figure 2). The corresponding risk reduction was 65% combining high DASH diet adherence with no diuretic use (HR, 0.35; 95% CI, 0.30-0.41); however, the joint effects between DASH diet and alcohol consumption were less notable (Figure 2). The joint associations with the Mediterranean, AHEI, and Prudent diets showed similar trends (eFigures 2-4 in the Supplement). The joint associations of Western diet and these factors were notable in the opposite direction, with a nearly 3.5-fold higher risk of gout among overweight or obese women most adherent to Western diet, and 3-fold higher risk among those using diuretics (eFigure 5 in the Supplement).

In this large prospective cohort study with 34 years of follow-up, we found that, among women, greater adherence to multiple healthy eating patterns consistent with the 2020 to 2025 Dietary Guidelines for Americans,¹⁷ including DASH, Mediterranean, AHEI, and Prudent, were consistently and independently associated with a lower risk of incident gout, whereas the Western diet was associated with a higher risk. Furthermore, these findings largely persisted in subgroups according to diuretic use, BMI, and alcohol intake, all strong risk factors for gout whose frequencies, we found, differ substantially from men. Collectively, these findings corroborate with the serum urate effect of these diets in randomized trial analyses^25,26,29 and extend the previous prospective evidence for male gout³⁰ compared with female gout, with additional guideline-based healthy eating patterns.^17,18

Although the 4 healthy dietary patterns were well correlated and overlapped in many of their components (ie, emphasis on consumption of fruits, vegetables, and whole grains) and metabolomic markers,^46,47 differences in the scoring of certain components (those shown to impact serum urate levels or gout risk) likely accounted for the minor differences in their associations with female gout. For example, the DASH diet assigns positive scores for consumption of low-fat dairy products,^41,48 and negative scores for sugar-sweetened beverages,³⁹ whereas the Mediterranean does not. The DASH diet has proven blood pressure benefits,^19,49,50 whereas the Mediterranean diet is thought to be more beneficial in insulin resistance and diabetes.⁵¹ These guideline-based diets likely affect gout risk through 2 key shared mechanisms involving insulin resistance and inflammation. Improved insulin resistance has been associated with increased urate excretion and lowering of serum urate.^28,52-55 Indeed, extending earlier human physiologic studies,^52,56 a recent bidirectional Mendelian randomization analysis⁵⁷ found that hyperinsulinemia may promote hyperuricemia, and not the other way around. The DASH-style pattern, and potentially others, may also promote urate excretion (thereby lowering serum urate levels) by protecting against kidney function declines, as suggested by an earlier analysis of the NHS.⁵⁸ Furthermore, higher adherence to the Mediterranean diet was also associated with higher urinary urate excretion (as measured in 24-hour urine samples) in the current female cohort as well as the male health professionals’ cohort.⁵⁹

At the same time, only about 20% of women with prolonged hyperuricemia will develop clinically evident gout,⁶⁰ and diet may further affect gout risk by modulating inflammatory pathways,^61,62 a supposition supported by prior studies of the current female cohort, wherein the Western pattern was positively associated with levels of C-reactive protein (CRP) and E-selectin, whereas the Prudent pattern was inversely associated, even after adjusting for BMI, a potential causal intermediate.⁶³ Similarly, the AHEI and Mediterranean diet scores were inversely associated with these markers of systemic inflammation in women in the NHS, along with interleukin-6,⁶⁴ a key cytokine in gouty inflammation. Finally, inflammatory biomarkers (including fibrinogen, high-sensitivity CRP, and glycoprotein acetylation) explained the largest proportion of the reduction in risk of incident cardiovascular events from the Mediterranean diet in the Women’s Health Study.²⁷

Regardless of the precise mechanisms, these findings show the 4 healthy dietary patterns, congruent with the latest Dietary Guidelines for Americans²⁸ with known cardiovascular-metabolic health benefits, also have the potential to simultaneously lower the risk of gout among women. These potential dual benefits contrast with the low-purine (low-protein) diet individuals with hyperuricemia or gout have traditionally been encouraged to follow,²⁸ which carries limited efficacy and an unfavorable cardiometabolic profile by indiscriminately replacing protein with carbohydrates (including refined) and fat (including unhealthy saturated or trans fats), which the 2020 to 2025 Dietary Guidelines discourage.¹⁷ Randomized clinical trials have demonstrated that the isocaloric DASH diets (without weight change), as well as Mediterranean or other weight loss diets can lower serum urate levels^25,26,29; consumption of individual food items,^39,41,65 and adherence to the DASH and Western dietary patterns,³⁰ have been associated with the risk of clinical gout, even after adjusting for BMI, but these exposures were studied primarily among men. Several cross-sectional analyses also reported consistent associations with serum urate levels or odds of hyperuricemia, independent of obesity.^32,66-68

The identification of multiple patterns of eating that can similarly reduce a woman’s risk of incident gout in our study allows more choice for potential personalization of dietary recommendations according to culinary traditions and personal preferences (including a genetic predisposition component⁶⁹), to enhance adherence, as well as comorbidity status (eg, DASH pattern for women with existing hypertension). It is also key to not consider diet and adiposity in isolation from one another, with our subgroup analysis (Table 3) suggesting diet may have less of an effect on gout risk among women who are overweight or obese (BMI ≥ 25) than women with normal body weight. At the same time, our exposure of interest was the isocaloric dietary pattern (ie, quality/types of foods), independent of calories (quantity of foods), which would directly affect adiposity and the risk of obesity, and serum urate levels correspondingly. Indeed, when combined with BMI in our joint association analysis we observed a risk reduction as high as 68% (HR, 0.32) (Figure 2), suggesting the combination of weight loss and adherence to a healthy guideline-based diet could reduce female gout risk even more substantially.

The potential pleiotropic benefits of these guideline-based cardiometabolic diets are highly relevant given the rising trajectory of gout disease burden among women, as described in recent analyses of the Global Burden of Diseases study,^7,8 and excess cardiometabolic comorbidity burden that has been documented disproportionately among women with gout, including higher prevalences of myocardial infarction, diabetes, and renal disease.^10,14 Therefore, lifestyle interventions that can simultaneously reduce the risk of female gout and cardiovascular disease, both of which have a tendency to be underrecognized and undertreated among women,^15,16 are of substantial importance.

A major strength of this study is the large number of cases of incident gout documented exclusively among women, and the prospectively collected and validated health and dietary information collected repeatedly over a long follow-up period. This study design has clear advantages over cross-sectional or case-control study designs often employed in initial lifestyle and nutritional studies.⁴⁴ A major weakness of cross-sectional studies^31,32,70 in this context is the lack of temporal sequence between dietary intake and gout, leaving the findings open to dietary change being a result of gout, whereas case-control studies are open to recall bias.⁷¹ Whereas future studies could investigate the effect of changes in adherence to healthy dietary patterns and gout risk, we assessed cumulative average intake using repeated dietary assessments, which accounted for variation in dietary consumption over time, and minimized measurement error. Residual measurement errors and misclassifications are expected to bias the results toward the null. Furthermore, the validity of the responses on the FFQ in subsamples of this population have been reported on multiple occasions.^33,34

As with any observational study, these findings are subject to unmeasured or residual confounding, though studying this cohort of female nurses minimizes potential for confounding associated with socioeconomic status, and we were able to obtain high-quality data with minimal loss to follow-up over 34 years. Although the absolute rates of gout and the distribution of dietary intake may not be representative of a random sample of American women, the biological effects of the dietary patterns on gout (as reflected by the HRs) should be similar. Indeed, because the risk factor burden for gout and its incidence tend to be higher in the general population, and higher among African Americans,⁷² the magnitude of the absolute risk difference associated with these diets is generally expected to be larger than that observed in the current study. Nevertheless, future research could examine the population contributions of diets and other risk factors for incident female gout, as done in men.⁴⁵

These large-scale, long-term prospective cohort findings extend the pleotropic benefits of the 2020 to 2025 Dietary Guidelines for Americans to female gout prevention, offering multiple healthy eating patterns that can be adapted to individual food traditions and preferences to reduce women’s underrecognized risk of developing gout while simultaneously addressing cardiovascular comorbidities.

Accepted for Publication: November 2, 2021.

Published Online: January 31, 2022. doi:10.1001/jamainternmed.2021.7419

Corresponding Author: Natalie McCormick, PhD, Harvard Medical School, Clinical Epidemiology Program, Division of Rheumatology, Allergy, and Immunology, Massachusetts General Hospital, 100 Cambridge St, Ste 1600, Boston, MA 02114 (nmccormick@mgh.harvard.edu).

Author Contributions: Dr Choi had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Drs Yokose and McCormick contributed equally to this work.

Concept and design: Yokose, Lu, Choi.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Yokose, McCormick, Lu, Choi.

Critical revision of the manuscript for important intellectual content: McCormick, Lu, Joshi, Curhan, Choi.

Statistical analysis: Lu, Joshi, Choi.

Obtained funding: Curhan, Choi.

Administrative, technical, or material support: Curhan, Choi.

Supervision: Curhan, Choi.

Conflict of Interest Disclosures: Dr Yokose reported grants from National Institutes of Health (NIH) [T32 AR007258] and grants from Rheumatology Research Foundation Scientist Development Award during the conduct of the study. Dr McCormick reported grants from Canadian Institutes of Health Research during the conduct of the study. Dr Curhan reported personal fees from OM1, Inc Employee, personal fees from UpToDate Royalties for my role as Section Editor and author, other from Allena Pharmaceuticals stock options for consulting, and grants from GlaxoSmithKline co-investigator; study of health effects of shingles outside the submitted work. Dr Choi reported grants from Horizon On an unrelated research project and personal fees from Allena Advisory Board outside the submitted work. No other disclosures were reported.

Funding/Support: This study was supported by the National Institutes of Health (NIH) [P50 AR060772, R01 AR065944, and UM1 CA186107]. Dr Yokose is supported by the NIH Ruth L. Kirschstein is supprted by the Institutional National Research Service Award [T32 AR007258] and the Rheumatology Research Foundation Scientist Development Award. Dr McCormick is supported by a Fellowship Award from the Canadian Institutes of Health Research. Dr Curhan is supported by the NIH [K24 DK091417].

Role of the Funder/Sponsor: The NIH had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.