Could a rural lifestyle decrease the prevalence of erectile dysfunction?


Joaquim de Almeida Claro, Division of Urology, Federal University of São Paulo, Paulista School of Medicine, Brazil. e-mail:



To determine the prevalence of erectile dysfunction (ED) in a specific population and explore potential correlates with lifestyle.


This prospective observational study, covering a population of a very small rural town, included 2000 men aged ≥ 20 years from a total population of 121 831 (51% female and 49% male). The International Index of Erectile Function was completed by each of the 2000 men at their homes over a 1-year period. Another questionnaire assessing socio-economic status and health-related determinants of ED were also completed.


All 2000 men completed the questionnaires; overall, only 34 reported ED (1.7%). The frequency of mild, mild to moderate, moderate and severe ED was 12%, 29%, 20% and 38%, respectively. Significantly more men aged > 51 years had ED than those aged <41 years (0.05% and 0.45%, respectively; P < 0.001). There was no difference in ED with salary levels.


The prevalence of ED in this particular rural population of Brazil was very low, at only 1.7%. Although ED increases with age, this association was not apparent for all age groups. It seems that several others factors, e.g. lifestyle, culture and diet, could be important for the onset of ED.


erectile dysfunction


International Index of Erectile Function.


Erectile dysfunction (ED) is a widespread and very common health problem affecting the quality of life of men of all ages [1,2]. According to the Massachusetts Male Aging Study projections, ED could affect up to 18 million men, and its current prevalence is 10–50% in the USA [3]. In Brazil, data from a large randomly sampled population-based survey indicated that 46% of men reported ED [4]. However, the information on the prevalence, incidence and risk factors for ED provided by all previous studies was collected in large metropolitan centres around the world. Because of this, the role of urbanization and quality of life in large cities on the prevalence of ED has not yet been investigated. In the present study we analysed data collected from a small rural town in Brazil.


The study was prospective, population-based, observational and conducted in a small town, Poços de Caldas, in south-eastern Brazil (533 km2, 121 831 inhabitants; males 56 977, 49%; females 60 117, 51%); 37 309 men were aged ≥ 20 years (63% of the males) and this sample was the target population. The sample selection comprised 2000 men and they were interviewed by one researcher only, complying with the following standards: (i) the town was divided into four zones (north, south, east and west); (ii) 25% of the town counties, corresponding to each zone, were selected for study; (iii) the population assessed was selected by random allocation of the streets of each county. In an allocated street, all houses were visited and all men aged ≥ 20 years were interviewed; (iv) all interviews were conducted by the same investigator, in a personal and confidential manner with the men, by themselves; (v) the International Index of Erectile Function (IIEF) [5] was administered by a urologist to each of the 2000 subjects. All participants provided written informed consent, and all data were collected by self-report only, at the subjects’ homes. Apart from the IIEF, another questionnaire assessing socio-economic status and health-related determinants of ED was also completed. The men were classified as having a specific disease if they reported receiving a physician’s diagnosis of that disease, or if they were taking medications for the disease of interest.

A descriptive analysis was used initially, calculating summary measures (mean, median, trends) [6]; the chi-square test was used to verify independence (with Yates’ correction when required) between the variables [7], with the comparison test of K proportions when necessary for comparing two proportions [6]; in all tests < 0.05 was taken as the rejection level for the null hypothesis.


All 2000 men completed both the IIEF and the other questionnaire. The mean (sd) of the sample was 33.2 (13.5) years, the oldest man being 76 years old. The results for age and the prevalence of ED is shown in Table 1.

Table 1. The distribution of ED according to age and salary
VariableWith ED, nNo ED, n% ED
Age, years
 20–30 1 4890.2
 31–40 1 4490.22
 41–50 5 4751.0
 51–60 9 3132.8
 >6118 2407.0
Salary, USA $
 0–75021 9612.2
 750–1500 8 5991.3
 1500–2250 1 1560.7
 2250–3000 1 1150.85
 >3000 3 1352.2

The participants had a higher than average level of education for Brazilian men; all had finished elementary school, and the income of the population is also shown in Table 1. Overall, only 34 men (1.7%) reported some degree of ED; the frequency of mild, mild to moderate, moderate and severe ED was 12%, 29%, 20% and 38%, respectively.

Although the prevalence of ED increased with age, from 0.05% in men aged 20–31 years to 0.90% in men aged > 60 years, this association was not always statistically significant. Significantly more men aged > 51 years had ED than men aged <41 years, at 0.05% and 0.45%, respectively (P < 0.001). There was no difference in the prevalence of ED with salary level.


It is estimated that > 25 million men aged 40–70 years will be affected by ED in the USA during the next year [8], and its prevalence could be 10–50%[3]. A similar prevalence of ED was reported in Brazilian studies, at 40–46%[2,4,9], but lower prevalence rates were reported in Germany, England and France, at ≈ 19%[10–12]. Furthermore nearly all these studies included a male population aged 40–70 years and were conducted in large cities or their neighbourhoods.

The present study was conducted in a randomly sampled population of men aged ≥ 20 years in a small town in rural Brazil. The high level of education of the participants could be explained by cultural issues and easy access to school in this region (both primary and high school). The prevalence of ED in these 2000 men (1.7%) is much lower than that previously reported in Brazil [2,4,9] and around the world [3,8,10–12].

That all the interviews were conducted by same investigator, which assured the same method for all participants, could be a source bias in some circumstances. However, in Brazilian small towns physicians are still respected in the community; people usually treat and confide in them as they would a priest or preacher. Furthermore, the specific doctor had a statement from our university that made clear the scientific and confidential character of the research. Thus, despite a few men possibly feeling uncomfortable, most could understand the level of the research and were comfortable with the interview.

Although the participants were younger than those in previous studies, this was probably not the only cause for the low prevalence of ED. Possibly the impressively low prevalence rate of ED is related to the particular lifestyle of people in this region of Brazil. The town is remote from any large city, life is still very calm, with dairy products and fruit forming the basis of the diet.

The Massachusetts Male Aging Study was conducted not just in an urban area of Massachusetts, but the results differed from those in the present study, although assessing the effects of many variables (genes, climate, pollution, diet, lifestyle and psychology) is daunting. What is selected for inclusion or exclusion in a study can lead to significantly different findings and to different conclusions. There is no easy way to devise representative sampling of a population or to confirm a cause-and-effect relationship. The most effective way to identify if a specific factor leads to a specific outcome is by comparing one randomly selected population to another, but it is impossible to assign traits randomly for factors such as climate, genes and diet. As a result, we compensated by using statistical methods.

We do not know yet if the specific characteristics of the selected area, including genes, lifestyle, climate, diet and psychology, are the main factors responsible for the much lower prevalence of ED. To answer this and other questions, other studies are in progress in similar Brazilian regions, with a deeper study of the specific population. More studies focused on specific populations are needed, but we assume that the prevalence of ED might be dissimilar among men with major differences in lifestyle, as in the present series and others.

In conclusion, ED, a common condition worldwide and with several important risk factors, might be affected by other factors, e.g. lifestyle, culture and diet, that could be important for the onset of ED. Future research should investigate specific populations to determine possible lifestyles that could help to preserve normal erectile function.


None declared.