To determine the relationship between popular diets and erectile function we queried the National Health and Nutrition Examination Survey, a cross-sectional dataset, between 2001 and 2004. All men aged 18–85 who answered the prostate and dietary questionnaires were included. Diets were categorized as Mediterranean, low-fat, low-carbohydrate, or nonrestrictive. Multivariable models were created to determine the relationship between erectile function and each diet. Among 4027 men, 649 (16.1%) met criteria for a low-fat diet, 1085 (26.9%) for a Mediterranean diet, and 0 (0%) for a low-carbohydrate diet. 1999 men (49.6%) had some degree of erectile dysfunction. Men with nonrestrictive diets were more likely to endorse normal erectile function compared with those adhering to the Mediterranean or low-fat diets (both p < 0.05) on univariable analysis. Multivariable analysis controlling for age, comorbidities, activity level, and body mass index showed no differences in erectile function among men adhering to a low-fat, Mediterranean diet, or nonrestrictive diet. There was no association between specific diets and erectile function. While additional prospective research is required to corroborate these findings, these data support the notion that individualized diets should be tailored toward goals of weight loss and reduction of comorbidity.
Erectile dysfunction (ED) is a common condition that affects up to 40% of men by the age of 70, and roughly 30 million men in the United States (US) [1, 2]. ED can have a detrimental impact upon a wide range of psychosocial domains, interpersonal relationships, and overall mental health . Furthermore, ED may serve as a harbinger of serious adverse cardiovascular events [4, 5]. The intimate relationship between cardiovascular morbidity and ED highlights the overlapping, modifiable risk factors between these two entities: diabetes, hyperlipidemia, sedentary lifestyle, and smoking [5, 6].
The recently published American Urological Association Guidelines on ED recommend a multifaceted, patient-tailored treatment approach that includes lifestyle modification . Lifestyle modification such as smoking cessation has reproducibly shown a decrease in ED [5, 7]. However, the effects of diet and exercise are somewhat unclear. The downstream impact of these interventions such as weight loss and subsequent reduction of comorbidity appear to impact erectile function, yet the specific effect of dietary composition remains unknown [5, 8,9,10,11]. Furthermore, many of the studies examining the relationship between diet and erectile function have been subject to criticism due to concerns regarding generalizability of findings—many examined very specific populations such as overweight men, type II diabetics, and men with ischemic heart disease, and the findings from these cohorts may not be applicable to the general population [5, 8,9,10,11].
With the increased emphasis on the importance of living a healthy lifestyle, ~45 million Americans will try a new diet in the coming year . Given the heterogeneity and limitations of prior studies analyzing the effects of popular diets on erectile function, we sought to examine the impact of the Mediterranean, low-fat, and low-carbohydrate diets on erectile function using a nationally-representative patient sample [13,14,15,16].
Materials and methods
Data source and study population
We utilized the National Health and Nutrition Examination Survey (NHANES) to examine the association between diet and erectile function. The NHANES is a cyclical, cross-sectional data set designed by the Center for Disease Control and Prevention (CDC) to investigate the general health of the US population . Data collected include patient demographics, surveys, medical comorbidities, and laboratory studies. In accordance with the National Center for Health Statistics (NCHS) Ethics Review Board, all participants submitted a written consent. This study received an exemption by the Northwestern University Fienberg School of Medicine Institutional Review Board. We utilized only those data cycles that included the full 2-day dietary interview and the prostate questionnaire (2001–2004). All men aged 18–85 years who completed both the dietary and ED instruments were included.
Assessment of dietary Intake
The validity and applicability of the NHANES dietary questionnaire has been previously described [17,18,19]. We used these data to determine the macronutrient intake for each patient during a two-day period based upon self-reported food consumption. Average macronutrient composition during the 2-day period was then utilized to characterize each patient as conforming to one or more of three common diets: the low-carbohydrate diet, low-fat diet, and the Mediterranean diet [13,14,15,16]. These diets were selected based on their safety and efficacy previously established by the Dietary Intervention Randomized Controlled Trial group . In brief, men met criterion for a low carbohydrate diet if they consumed ≤20 g of carbohydrates daily. Men met criterion for a low-fat diet if they consumed ≤1800 kCal per day, of which no more than 30% of calories were from fat and no more than 10% of calories were from saturated fat, and took in <300 mg of cholesterol daily. Last, men met criterion for a Mediterranean diet if they consumed ≤1800 kCal, of which no more than 40% of calories were from fat (without any restriction for saturated fat or cholesterol).
Men who did not meet criteria for one of the aforementioned diets were considered to have a nonrestrictive diet. Since the restrictive diets were not mutually exclusive (i.e., an individual could meet criteria for both a low-fat and Mediterranean diet, simultaneously), diet comparisons were made only between each restrictive diet and the nonrestrictive diet group, as comparison of restrictive diets would result in a subset of patients included in both comparison groups.
Given the inherent limitations of self-reported dietary intake, multiple checks were utilized to optimize integrity of dietary data. NHANES performed an internal quality control for assessment of dietary recall status, and only men whose dietary recall status was qualified as “reliable” were included in the analysis. In addition, the dietary interview included a question, “Was the amount of food that you ate yesterday much more than usual, usual, or much less than usual?” To ensure that the recorded nutritional intake was representative of an individual’s typical diet, only those participants who replied “usual” were included.
Primary outcome and covariables
The primary outcome was self-reported erectile function. NHANES assessed erectile function with a single question, “Many men experience problems with sexual intercourse. How would you describe your ability to get and keep an erection adequate for satisfactory intercourse?” Responses were registered using a Likert scale with the following options: always or almost always able, usually able, sometimes able, and never able. For all statistical analysis, responses were coded according to a one to four scale (one being always or almost always able and four being never).
Demographic variables such as age, race, income, and education level were examined. To minimize confounding, covariables such as comorbidities, substance use, body mass index (BMI), activity level, history of prostate cancer, sex hormone binding globulin, albumin, and testosterone levels were also analyzed. Comorbidities were further characterized using a previously established NHANES-specific comorbidity index which includes arthritis, asthma, cancer, congestive heart failure, myocardial infarction, and stroke . While diabetes and metabolic syndrome were included in the original comorbidity index, we classified these conditions as independent variables for the present study due to their profound impact upon erectile function.
Participant activity levels were determined based on responses to two questions: “Over the past 30 days, did you do any vigorous activities for at least 10 min that caused heavy sweating, or large increases in breathing or heart rate?” and “Over the past 30 days, did you do moderate activities for at least 10 min that cause only light sweating or a slight to moderate increase in breathing or heart rate?” Based upon positive and negative responses to these questions, baseline activity levels were characterized as vigorous, moderate, or less than moderate.
All laboratory values were measured using validated techniques described in detail by the CDC [21, 22]. Morning serum testosterone was measured using a validated isotope dilution liquid chromatography tandem mass spectrometry method .
We used multiple analytic approaches, treating erectile function as both a continuous and categorical variable, utilizing multiple different cut-off points for the latter. To exhaustively investigate the aforementioned association we performed univariable and multivariable analysis on all logical permutations of participant responses. We compared men who were always or almost always able to get erections (Likert score of 1) to men who were not always able to get an erection (Likert score of 2,3 or 4). Next, we compared men who were always or usually able to get an erection (Likert score of 1 or 2) to men who were sometimes or never able to get an erection (Likert score 3 or 4). Last, we compared men who were always, usually or sometimes able to get an erection (Likert score of 1, 2, or 3) to men who were never able to get an erection (Likert score of 4).
All statistical analysis was performed using SAS 9.4 (SAS Inc., Cary, NC). The 2001–2002 and 2003–2004 data sets were combined using respondent sequence numbers. As directed by the NCHS, we used the appropriate the strata, cluster, and sample weights to account for NHANES’ complex design .
For comparison of demographic characteristics among diet groups, we used both chi-squared and t tests, where applicable. Univariable regressions were performed to determine the effects of individual covariables on erectile function. Covariables that were significant on univariable analysis were then used to construct a multivariable regression. Odds ratios (OR) and their respective 95% confidence intervals (CI) were calculated and statistical significance was set to p < 0.05 (two sided).
Diet and demographic data
Among 4027 men that met inclusion criteria, 2942 (73.1%) men did not endorse having a restrictive diet. The most common restrictive diet was the Mediterranean diet (N = 1085, 26.9%) followed by the low-fat diet (N = 649, 16.1%). No participants met the strict criteria for a low-carbohydrate diet.
Demographic data for the entire cohort are presented in Table 1. Men adhering to the Mediterranean or low-fat diet were more likely to have one or more comorbidity (53.19% and 51.77%, respectively) compared with men on nonrestrictive diets (38.41%, p < 0.001 for both). Furthermore, men on nonrestrictive diets were less likely to have diabetes (8.18%) vs. those adhering to the Mediterranean (16.67%) or low-fat diet (15.35%) (p < 0.001 for both). There were no significant differences in BMI, alcohol use, activity level, and prostate cancer status between diet groups.
Demographic and clinical characteristics
Erectile function for the cohort is displayed in Table 2. The median age of men always able to obtain an erection was 38 years (Interquartile Range [IQR] 29–47) vs. 55 years (IQR 43–68) for men not always able to achieve an erection (p < 0.001). Similarly, there was a significant difference in age when comparing men with some degree of erectile function vs. none (median 42 [IQR 31–53] vs. 72 [IQR 61–78] years respectively, p < 0.001). Men endorsing no erectile function were more likely to have low testosterone (≤300 ng/dL) compared with men with any erectile function (41.3% vs. 12.37%, p < 0.001). In all comparisons, diabetic men were less likely to endorse erections compared with nondiabetic men. The largest difference was detected when comparing the ability to always achieve vs. not always achieve an erection (3.79% diabetic vs. 17.62% nondiabetic, p < 0.001).
Effect of diet upon erectile function
Univariable analysis revealed that men with nonrestrictive diets were more likely to endorse better erectile function compared with men adhering to the Mediterranean or low-fat diets ([OR] = 0.341, 95% [CI] 0.266–0.436; OR 0.476, 95% CI 0.379–0.597, both p < 0.001) when erectile function was analyzed as a continuous variable (Table 3). This finding was preserved on univariable analysis when considering erectile function as a categorical variable in a variety of permutations (Table 2).
When controlling for confounders including age, testosterone level, BMI, smoking status, diabetes, and other comorbidities on multivariate analysis, no significant difference in erectile function was found between men adhering to the restrictive diets compared with men on nonrestrictive diets. When considering erectile function as a continuous variable, men adhering to the low-fat diet did have better erectile function (OR 0.383, CI 1.32–1.109, p = 0.075), which approached but did not reach statistical significance (Table 4). Supplementary Table 1 shows additional multivariable models examining erectile function as a categorical variable. None of these models demonstrated a difference in erectile function between diet groups.
As the US strives to be a healthier nation, it is important for clinicians to understand how weight loss strategies such as popular diets affect common medical conditions. While the use of restrictive diets has been shown to be an effective weight loss strategy, restrictive diets’ effects on ED are heterogeneous [13, 25]. Multiple randomized prospective trials have analyzed the effects of diet on erectile function [8, 26,27,28,29]. Esposito et al. demonstrated that obese men with ED were more likely to regain normal erectile function by increasing their activity level and reducing their caloric, carbohydrate, and fat intake (31% vs. 6%, p = 0.001) . The same group also found that adherence to a Mediterranean diet in men with ED and metabolic syndrome improved international index of erectile function (IIEF) scores by an average of three points compared with controls (p = 0.01) . Similarly, Khoo et al. found significant improvements in IIEF scores in obese, diabetic males who lost weight consuming a low-calorie, low-fat diet .
While these studies demonstrated a strong relationship between diet and erectile function, the study designs render it impossible to determine whether the effects of diet on erectile function were attributable to dietary composition or other collinear outcomes such as weight loss and reduction of medical comorbidity. Indeed, a recent comprehensive review concluded that improvement in erectile function among overweight men adhering to a low-fat diet was likely the result of weight loss . Similarly, a randomized controlled trial with over eight years follow-up showed that men adhering to the Mediterranean diet showed slower deterioration of IIEF scores compared with men adhering to the low-fat diet by 1.16 points (p = 0.024), but they also experienced more weight loss by nearly 1 kg as well a reduction in hemoglobin A1c by 0.3% . The individuals in this study adhering to the Mediterranean diet showed a larger reduction in systolic blood pressure by 1.8 mm of mercury, and multivariable analysis did not control for BMI . While these studies help to demonstrate that improvements in cardiovascular health parameters can improve erectile function (or slow its decline), it can be difficult to determine if there is a direct impact of dietary composition on erectile function.
Basic science studies corroborate these findings. Animal models have demonstrated changes in diet to be associated with alterations in corporal bodies [30,31,32]. Mice fed a high energy diet (compared with those fed a low energy diet) saw an increase in type I collagen deposition in their cavernosal bodies as well as a decrease in smooth muscle fibers, a pattern often seen in erectile dysfunction . Another study demonstrated a decreased number of erections and increased latency period of apomorphine-induced erections in mice . However, none of these animal studies analyzed changes in erectile function or corporal compositions independent of weight gain. In fact, across all studies, animals fed higher calorie diets gained significantly more weight, had higher blood pressure and worse lipid profiles than those eating a normal calorie diet, rendering it difficult to distinguish the impact of nutritional intake from these other potential mechanisms for the physiologic findings.
We designed our study to address some of the aforementioned limitations. This is the largest analysis of the association between macronutrient restrictive diets and erectile function in the US. By controlling for BMI, diabetes, comorbidity, and testosterone levels in multivariable models, we were able to examine the independent effect of diet on erectile function. Furthermore, given the heterogeneous and population-based composition of the cohort, our results are generalizable to the US population.
Neither the low-fat nor the Mediterranean diet was independently associated with improved erectile function when compared with a nonrestrictive diet. This does not refute the overall health benefits of previously mentioned studies. However, the current data demonstrate the collinearity among weight loss, improved exercise tolerance, improved glucose tolerance, and erectile function, which is the likely driver of differences in erectile function noted in the aforementioned studies. In lieu of a direct effect of dietary composition on erectile function, diet more likely results in the indirect improvement of erectile function through reduction in comorbidity.
These data must be viewed within the context of their limitations and the study design. First, this was a cross-sectional study due to the inherent data collection strategy of NHANES. As such, there was no longitudinal follow-up, and changes in diet, weight, and erectile function could not be evaluated over time, rendering it impossible to determine causality. Second, NHANES does not record IIEF score, and the erectile function questionnaire utilized by NHANES was not previously validated. To address this limitation, we performed multiple sensitivity analyses treating erectile function as both a continuous and categorical variable, and the results were stable across all approaches. Third, due to the cross-sectional nature and study design, there was no mechanism to determine whether men were truly adherent to their self-reported diet, which was also subject to recall bias. Last, given the strict definition of a low-carbohydrate diet as well as the cross-sectional nature of this data set, we were only able to assess men initiating a ketogenic diet and therefore may not have captured men who were already ketogenic and maintaining a low-carbohydrate diet.
In a nationally representative sample, when controlling for BMI and medical comorbidities, there was no difference in erectile function between men adhering to restrictive diets and men on nonrestrictive diets. However, potential benefits of restrictive diets on BMI, comorbidities and glucose tolerance may indirectly impact erectile function. Future prospective research is required to validate these findings.
Lewis RW, Fugl-Meyer KS, Corona G, Hayes RD, Laumann EO, Moreira ED. et al. Definitions/epidemiology/risk factors for sexual dysfunction. J Sex Med. 2010;7:1598–607.
McKinlay JB. The worldwide prevalence and epidemiology of erectile dysfunction. Int J Impot Res. 2000;12 Suppl 4:S6–11.
Latini DM, Penson DF, Wallace KL, Lubeck DP, Lue TF. Clinical and psychosocial characteristics of men with erectile dysfunction: baseline data from ExCEED. J Sex Med. 2006;3:1059–67.
Lee JH, Ngengwe R, Jones P, Tang F, O’Keefe JH. Erectile dysfunction as a coronary artery disease risk equivalent. J Nucl Cardiol. 2008;15:800–3.
Burnett AL, Nehra A, Breau RH, Culkin DJ, Faraday MM, Hakim LS, et al. Erectile dysfunction: AUA guideline. J Urol. 2018;200:633–41.
Martin-Morales A, Sanchez-Cruz JJ, Saenz de Tejada I, Rodriguez-Vela L, Jimenez-Cruz JF, Burgos-Rodriguez R. Prevalence and independent risk factors for erectile dysfunction in Spain: results of the Epidemiologia de la Disfuncion Erectil Masculina Study. J Urol. 2001;166:569–74. discussion 574-575
Kovac JR, Labbate C, Ramasamy R, Tang D, Lipshultz LI. Effects of cigarette smoking on erectile dysfunction. Andrologia. 2015;47:1087–92.
Esposito K, Giugliano F, Di Palo C, Giugliano G, Marfella R, D’Andrea F, et al. Effect of lifestyle changes on erectile dysfunction in obese men: a randomized controlled trial. JAMA. 2004;291:2978–84.
Esposito K, Ciotola M, Giugliano F, De Sio M, Giugliano G, D’armiento M, et al. Mediterranean diet improves erectile function in subjects with the metabolic syndrome. Int J Impot Res. 2006;18:405–10.
Khoo J, Piantadosi C, Worthley S, Wittert GA. Effects of a low-energy diet on sexual function and lower urinary tract symptoms in obese men. Int J Obes. 2010;34:1396–403.
Kałka D, Domagała Z, Dworak J, Womperski K, Rusiecki L, Marciniak W, et al. Association between physical exercise and quality of erection in men with ischaemic heart disease and erectile dysfunction subjected to physical training. Kardiol Pol. 2013;71:573–80.
Searing L. The Big Number: 45 million Americans go on a diet each year. The Washington Post [Internet]. 2018 Jan 1[cited 2020 May 26]. Available from: https://www.washingtonpost.com/national/health-science/the-big-number-45-million-americans-go-on-a-diet-each-year/2017/12/29/04089aec-ebdd-11e7-b698-91d4e35920a3_story.html.
Shai I, Schwarzfuchs D, Henkin Y, Shahar DR, Witkow S, Greenberg I, et al. Weight loss with a low-carbohydrate, Mediterranean, or low-fat diet. N Engl J Med. 2008;359:229–41.
Krauss RM, Eckel RH, Howard B, Appel LJ, Daniels SR, Deckelbaum RJ, et al. AHA Dietary Guidelines: revision 2000: a statement for healthcare professionals from the Nutrition Committee of the American Heart Association. Stroke. 2000;31:2751–66.
WILLETT W EAT, DRINK, AND BE HEALTHY: the harvard medical school guide to healthy eating. Place of publication not identified: FREE Press; New York, NY; 2017.
Atkins RC. Dr. Atkins’ new diet revolution. 1st Avon pbk. ed. New York: Avon Books; 2002. 540 p.
Ahluwalia N, Dwyer J, Terry A, Moshfegh A, Johnson C. Update on NHANES dietary data: focus on collection, release, analytical considerations, and uses to inform public policy. Adv Nutr. 2016;7:121–34.
Dwyer J, Picciano MF, Raiten DJ, Members of the Steering Committee. Estimation of Usual Intakes: What We Eat in America–NHANES. J Nutr. 2003;133:609S–23S.
Raper N, Perloff B, Ingwersen L, Steinfeldt L, Anand J. An overview of USDA’s dietary intake data system. J Food Composition Anal. 2004;17:545–55.
He Z, Charness N, Bian J, Hogan WR Assessing the comorbidity gap between clinical studies and prevalence in elderly patient populations. In: 2016 IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI) [Internet]. Las Vegas, NV, USA: IEEE; 2016 [cited 7 Dec 2018]. p. 136–9. http://ieeexplore.ieee.org/document/7455853/.
NHANES 2001–2002 Laboratory Methods [Internet]. CDC, National Center for Health Statistics; [cited 20 Mar 2019]. https://wwwn.cdc.gov/nchs/nhanes/continuousnhanes/labmethods.aspx?BeginYear=2001.
NHANES 2003–2004 Laboratory Methods [Internet]. CDC, National Center for Health Statistics; [cited 20 Mar 2019]. https://wwwn.cdc.gov/nchs/nhanes/continuousnhanes/labmethods.aspx?BeginYear=2001.
Pirkle J. Laboratory Procedure Manual: Total Testosterone [Internet]. CDC; 2011-2012 [cited 2020 May 26]. Available from: https://www.cdc.gov/nchs/data/nhanes/nhanes_11_12/tst_g_met.pdf.
Mirel LB, Mohadjer LK, Dohrmann SM, Clark J, Burt VL, Johnson CL, et al. National Health and Nutrition Examination Survey: estimation procedures, 2007–2010. Vital Health Stat. 2013;2:1–17.
La J, Roberts NH, Yafi FA. Diet and Men’s Sexual Health. Sex Med Rev.2018;6:54–68.
Wing RR, Rosen RC, Fava JL, Bahnson J, Brancati F, Gendrano Iii INC, et al. Effects of weight loss intervention on erectile function in older men with type 2 diabetes in the Look AHEAD trial. J Sex Med. 2010;7:156–65.
Khoo J, Piantadosi C, Duncan R, Worthley SG, Jenkins A, Noakes M, et al. Comparing effects of a low-energy diet and a high-protein low-fat diet on sexual and endothelial function, urinary tract symptoms, and inflammation in obese diabetic men. J Sex Med. 2011;8:2868–75.
Maiorino MI, Bellastella G, Caputo M, Castaldo F, Improta MR, Giugliano D, et al. Effects of Mediterranean diet on sexual function in people with newly diagnosed type 2 diabetes: The MÈDITA trial. J Diabetes Complicat. 2016;30:1519–24.
Moran LJ, Brinkworth GD, Martin S, et al. Long-term effects of a randomised controlled trial comparing high protein or high carbohydrate weight loss diets on testosterone, SHBG, erectile and urinary function in overweight and obese men. Shore N, editor. PLoS One. 2016;11:e0161297.
de Souza ILL, Barros BC, de Oliveira GA, Queiroga FR, Toscano LT, Silva AS, et al. Hypercaloric diet establishes erectile dysfunction in rat: mechanisms underlying the endothelial damage. Front Physiol. 2017;8:760.
Alves-Pereira JL, Frantz EDC, Pires LAS, Babinski MA, da Fonte, Ramos C. Effects of a high energy density diet in the “corpus cavernosum” of mice. Int J Impot Res. 2019;31:126–31.
Yeşilli Ç, Yaman Ö, Anafarta K. Effect of experimental hypercholesterolemia on cavernosal structures. Urology. 2001;57:1184–8.
Conflict of interest
The authors declare that they have no conflict of interest.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Fantus, R.J., Halpern, J.A., Chang, C. et al. The association of popular diets and erectile function among men in the United States. Int J Impot Res 33, 548–555 (2021). https://doi.org/10.1038/s41443-020-0313-x