Can Erectile Dysfunction be Delayed or Improved with the Mediterranean Diet?

A summary of Bauer SR, Breyer BN, Stampfer MJ, Rimm EB, Giovannucci EL, Kenfield SA. “Association of Diet with Erectile Dysfunction Among Men in the Health Professionals Follow-up Study:, JAMA Network Open.2020;3(11):e2021701. doi:10.1001/jamanetworkopen.2020.21701

Introduction

“Approximately 10% of men aged 40-70 have severe or complete erectile dysfunction (ED)… with an additional 25% … having moderate or intermittent erectile difficulty,” (Boston University, Epidemiology of ED). The onset of any level of ED can create stress, changes in self-esteem, and impact intimacy with partners. But what if you could delay or prevent ED with dietary changes?

Erectile dysfunction has a variety of causes. Physical conditions including Diabetes Type II (DT2), cardiovascular disease (CD), hypertension (high blood pressure), and hyperlipidemia (high cholesterol) are known correlates for ED. Many medications for psychological health, including selective norepinephrine/serotonin uptake inhibitors (SNRIs) are known to cause ED and other sexual dysfunction in people with penises (Trinchieri et al, 2021). Beyond the physical causes of ED, feelings of blame, anger, anxiety, bitterness, and depression are highly correlated with the presence of sexual dysfunction (Huri et al., 2016).

Finding methods which delay or decrease the chances of developing erectile dysfunction are of interest to medical researchers. Because ED is correlated with the presence of cardiovascular disease, DT2, hyperlipidemia, and inflammation markers, looking at common behavioral changes which may reduce these issues as a possible form of  prevention of ED makes sense. “Heart healthy” diets are a common suggestion provided to patients experience any of the above issues. In recent decades, the “Mediterranean diet”- a diet centered around vegetables, legumes, fish, whole grains, and olive oil- has been promoted as a way to reduce cholesterol, inflammation, reduce insulin spike, and improve overall health.

Bauer et al. Study Design

Scott R. Bauer, MD and Stacey A. Kenfield, ScD, primary investigators, used longitudinal data from the “Health Professional Follow-up Study” which began in 1986. Questionnaires were mailed to male health professionals in the US. The initial response included 51,529 returned questionnaires. The questionnaire included a section where men estimated the amount and frequency of eating 130 items (FFQ). These participants were mailed follow-up questionnaires every two years. Beginning in 2000, the questionnaire included a question asking about the ability to maintain an erection without treatment.

The study used data on erectile function from questionnaires administered between 2000 and 2014 and the FFQ from 1986-2014. Men who had not filled out the FFQ in 1989 (8,505), who had died (5,510), and those diagnosed with prostate, testicular, or bladder cancer prior to 1989 (445) were excluded. Additionally, any participant consuming less than 800 calories or more than 4,200 calories a day were excluded because the researchers judged this level of food intake “improbable.” Additionally, men with CD, DT2, or a history of stroke or heart attack were excluded (nearly 21% of the sample). The researchers justified the exclusion of these men based on the assumption they must have had a poor diet which resulted in the presence of these conditions.

As a result of the exclusions, 21,469 men were remaining in the ED study. Of those participants, 21,155 were white (98.5 percent).

Participants self-reported food intake was based on their estimates of how often and how much of 130 common American dietary foods they consumed in the prior year. Their estimated daily calories were calculated along with estimates of how many servings of legumes, fruit, vegetables, nuts, fish, red and/or processed meats, and alcohol they consumed. They were assigned points for eating more legumes, fruit, vegetables, nuts and fish and less red and/or processed meat and alcohol than recommended in the “Mediterranean diet.”

In 2010 the survey altered the food intake questionnaire and replaced adherence to the MD to adherence to the Healthy Eating Index. This measure similarly prioritized monosaturated fats, vegetables, fruits, and legumes.

Potential confounding variables including smoking, taking specific classes of medications, age, frequency of physical activity, and BMI were included in a multivariate analysis.

Findings

The authors found that men over 60 who scored in the highest on maintaining Mediterranean diet standards were less likely to develop ED. When they factored in smoking, BMI, and activity levels, adherence to the MD still proved significant in not developing ED.

Author Discussion

The authors believe that their study was the first to look at a large survey population and connect adherence to the MD to decreasing the probability of developing ED. They suggest that doctors council patients on the benefits of the diet as preventative care for sexual health.

They cite the MÉDITA study as providing limited evidence to further support their findings The MÉDITA study looked at 108 men at one location who were randomly assigned the MD or a low-fat diet. The group assigned to the MD (s=54) had fewer incidents of ED over the course of the study when compared to men on a low-fat diet.

This study differs from Bauer’s et al. in some significant ways. Firstly, the group in the MÉDITA study had DT2 where all men with type 2 diabetes were eliminated from the Bauer study. Second, the group in the MÉDITA study had a diet assigned to them. The men in the Bauer study did not have dietary guidelines imposed on them. Finally, the MEDITA participants were supposed to limit caloric intake to 1,800 a day. There is no report in the Bauer study of the average number of calories consumed by participants beyond the 800-4200 per day range.

Author Conclusions

The authors admit to some limitations of their study. They acknowledge that the men their study were not assigned to a diet. Additionally, the self-report allows for some bias to come into the study. They conclude that the use of the tools they chose would negate any bias which may have arisen.

While the authors recognize a few (minor) limitations to their study, they fail to account for significant limitations and confounding variables.

While recognizing their study may have a few limitations, the authors conclude that men who adhere to a Mediterranean Diet will experience less ED as they age.

Limitations of the Study

Lack of Racial Representation

The lack of racial representation is the first issue with this study. With nearly all participants being white men, any conclusions cannot be generalized to other racial groups. The exclusion of minorities is also confounded with the lack of economic data provided for the subjects.

Lack of Income Data

The American healthcare workforce is stratified by race and income. The Bureau of Labor Statistics (BLS) report that the highest paying medical professions (MDs, DOs) are majority white (over 60%) while the lowest paying professions (medical technicians, medical assistants) are predominantly nonwhite. This creates an enormous income gap between the highest and lowest paid healthcare occupations. In 2014, the annual average income for a nursing assistant was $26,250 while the average income for the top five paying healthcare positions averaged $104,180 (BLS, Highest and Lowest Paying Healthcare Occupations, 2014). Adjusted for inflation the gap in 2025 dollars is $36,698 to $104,167.69. [i]This gap creates significant differences in access to food, healthcare, and preventative services.

Exclusion of Significant Confounding Variables

The authors state they exclude any participant reporting heart attack, stroke, or Type II diabetes because assumed an “unhealthy eating pattern prior to their diagnosis.” This assumption cannot be justified by scientific research. Much of the current research on diet in relationship to development of health concerns, specifically obesity and diabetes, fails to address confounding variables. Diet is correlated with both culture and income. People who have lower incomes are less likely to access healthcare, afford or access sufficient Mediterranean diet foods to meet caloric needs, and more likely to experience healthcare barriers such as racism, language barriers, and locations near them.

Additionally, by eliminating minority men and not breaking out the sample by income levels, the researchers ignore psychological precursors to erectile dysfunction. There is significant documentation of depression and anxiety correlating with erectile dysfunction. Both the presence of these conditions and their common medical treatments increase the chance of experiencing ED. Black men are significantly more likely than non-Hispanic white men to report symptoms of depression in the past two week according to the CDC. Black men are significantly less likely to access mental healthcare when compared to the general adult population of the U.S.

The survey did not provide information about health care insurance. For a significant period of this study, health care was not a mandatory requirement. For the years 1998-2008 the percentage of uninsured Americans between 18-65 years old hovered around 15 percent (https://www.cbpp.org/research/number-and-percentage-of-americans-who-are-uninsured-climbs-again-in-2006)  Mental healthcare parity (if mental health was covered by an insurer, it must receive the same type of coverage as physical health) was not required in the U.S. until 2010. Requiring mental health coverage by insurers did not occur until 2014. Due to income and insurance disparities, it is difficult to determine if men who were paid on the lower end of the healthcare payment spectrum would have access to treat clinical depression and anxiety.

Choice of Dietary Measures

The survey asked the participants to report on the frequency they consumed 130 foods over the previous year. The use of self-reporting of both the consumption of various foods and the amount (servings) provides significant room for error. Additionally, the food report only included 130 common American dietary foods. There were no adjustments for regional or cultural diets.

The calculation of adherence to the Mediterranean diet was based on a points system. Points were assigned the self-report food intake questionnaire for consuming more than the “average recommended servings” of legumes, vegetables, fruits, nuts, grains, and fish. Additional points were given for people who consumed less than the “average recommended servings” of red or processed meats and less than one drink of alcohol per day.

The assumption behind this model is that all bodies need similar caloric intakes and macronutrients. The reality is, depending on your physical composition of muscle, fat, and other tissue; disability and/or presence of chronic pain; age; and genetic factors, the same foods cause different responses. For example, someone with Chron’s or Colitis should avoid nuts because they can become lodged in the intestines and cause pain and inflammation. This individual would not receive a point for consuming more than the “average recommended serving” of nuts in their diet. For the Chron’s patient, eliminating nuts helps reduce inflammation – a purported benefit of the Mediterranean diet – but would be assumed to have a “poorer” diet because of the avoidance.

The study also eliminated individuals consuming more than 4,200 calories per day because this level of intake was deemed “improbable.” Again, this is an assumption about all bodies performing the same way.

This study also fails to account for weight fluctuations during the survey period. While the use BMI as a measure of “health” they do not calculate for changes weight over time. Weight cycling is a known contributor to weight gain and inflammation. Both inflammation and weight gain are known contributors to ED.

Finally, subjects were assigned as single score for adherence to the MD. The authors believe this to be a benefit for the measure as it accounts for more than a decade of self-reports. However, this fails to account for dietary changes over more than a decade. Poor adherence to the MD during younger years and a more strict adherence in the years closest to reporting ED would appear no different statistically than strong adherence in the early years of the study and poor adherence in the proximal years.

The impact of dietary changes to meet different needs as the body ages or experiences a variety of illnesses cannot be calculated with the statistics presented in this study. Additionally, dietary changes which accompany increased income or access which correlates positively with age are not accounted for.

The best this study can conclude is that for white men over 60 who work in healthcare in the US and choose to (and can afford) to adhere to a MD and who have no history of cancer, Type 2 diabetes, or stroke may reduce their chance of developing ED by eating this type of diet.

References

Bauer SR, Breyer BN, Stampfer MJ, Rimm EB, Giovannucci EL, Kenfield SA. “Association of Diet with Erectile Dysfunction Among Men in the Health Professionals Follow-up Study:, JAMA Network Open.2020;3(11): e2021701. doi:10.1001/jamanetworkopen.2020.21701

Boston University Medical Center. ”Epidemiology of ED” webpage. Accessed 10 Feb 2026. https://www.bumc.bu.edu/sexualmedicine/physicianinformation/epidemiology-of-ed/

Center on Budget and Policy Priorities. “Report: Numbers and Percentage of American Who are Uninsured Climbs Again.” Revised Aug. 31, 2007. Webpage. Accessed 10 Feb 2026. https://www.cbpp.org/research/number-and-percentage-of-americans-who-are-uninsured-climbs-again-in-2006

Fayer, Stella and Audrey Watson. (2015). “Employment and Wages in Healthcare Occupations,” U.S. Bureau of Labor Statistics, Washington, DC. https://www.bls.gov/spotlight/2015/employment-and-wages-in-healthcare-occupations/home.htm

Huri HZ, Mat Sanusi ND, Razack AH, Mark R. Association of psychological factors, patients' knowledge, and management among patients with erectile dysfunction. Patient Prefer Adherence. 2016 May 13;10:807-23. doi: 10.2147/PPA.S99544. PMID: 27257374; PMCID: PMC4874731.

Maiorino, Maria Ida, Giuseppe Bellastella, Paolo Chiodini, Ornella Romano, Lorenzo Scappaticcio, Dario Giugliano, Katherine Esposito. : Primary Prevention of Sexual Dysfunction With Mediterranean Diet in Type 2 Diabetes: The MÈDITA Randomized Trial.: Diabetes Care 1 September 2016; 39 (9): e143–e144. https://doi.org/10.2337/dc16-0910

Office of Minority Health. 2026. “Mental Health in Black/African Americans.” Department of Health and Human Services, webpage. Accessed 10 Feb 2026. https://minorityhealth.hhs.gov/mental-and-behavioral-health-blackafrican-americans

Margherita Trinchieri MD , Martina Trinchieri MSc, Gianpaolo Perletti PhD , Vittorio Magri MD, Konstantinos Stamatiou MD, Tommaso Ca MD, Emanuele Montanari MD , Alberto Trinchieri MD.  (2021). ”Erectile and Ejaculatory Dysfunction Associated with the Use of Psychotropic Drugs: A Systematic Review,” The Journal of Sexual Medicine, 18(8):1354-1363.

https://doi.org/10.1016/j.jsxm.2021.05.016

Villarroel, Maria A., Ph.D., and Emily P. Terlizzi, M.P.H. (2020). ”Symptoms of Depression Among Adults, 2019.” National Center for Health Statistics, Data Brief 379, accessed 10 Feb 2026. https://www.cdc.gov/nchs/products/databriefs/db379.htm

[i] *This was calculated by using the inflation calculator here.