Aging male's symptoms in a Southern Brazil population: lifestyle effects after the age of 40

Abstract

Background. The interest of epidemiological research about male's aging increased in recent years along with the need to evaluate health-related quality of life. We conducted a population-based cross-sectional study to identify the prevalence of aging male's symptoms (AMS) and factors associated to this condition.

Methods. The study included 421 men aged 40 years or older, living in the urban area of Pelotas, Southern Brazil. The questionnaire covered sociodemographic, behavioral, and health variables, and to verify the aging male's symptoms, the AMS Scale was used.

Results. Moderate/severe male's symptoms was considered positive (AMS scores equal or above 37 points) in 20% of men (95% CI 16.1; 24.3). After controlling for confounders, the AMS was significantly associated with aging, self-perceived health status and smoking. The symptoms were more severe among physically inactive men.

Conclusions. Our findings support the necessity to maintain healthy behaviors like not smoking and regular physical activity since such changeable behaviors could reduce the AMS acceleration and increase life quality and expectancy.

Keywords:

• Aging
• male
• aging male's symptoms
• physical activity level and epidemiology

Introduction

Aging is an irreversible physiological process that affects men and women. The decreasing fecundity has been pointed as the main factor leading to demographic aging, although, in countries like Brazil a small decrease on mortality rates may contribute to the process as well [1,2]. The physiological aging is determined by genetic (DNA) and cell changes (free radicals and apoptosis) and by systemic, neuroendocrine, and immunologic alterations [3,4]. However, the pattern and severity of physiologic deterioration may be influenced, or even reverted by regular physical activity. Other behavioral characteristics related to lifestyle, such as smoking, alcohol ingestion, eating habits, and personality also influence on the advancement of the aging process [5].

With respect to aging and physical inactivity, a vicious circle has been demonstrated: aging leads to physical inactivity and physical inactivity speeds up the process by increasing physical problems like lack of fitness, muscle and bone frailty. The whole process results in lower self-esteem and motivation, increasing the chance of mental health problems, like anxiety and depression [6]. Among the main symptoms associated to men's aging are: decrease in strength and muscle mass; abdominal fat accumulation (especially central adiposity leading to insulin resistance and atherosclerosis); lowered sex drive, mineral bone and pubic hair; cognitive losses; depression, insomnia; sudoresis and a marked decrease in overall well being. Hence, these symptoms are psychological, somatic, and sexual [7,8].

The prevalence of aging symptoms in men has been studied, and the natural tendency is that the symptoms aggravate with age [9–11]. Japanese and German studies report symptoms prevalence rates of 22.7 and 18%, respectively [10,11]. According to Ichioka et al. [11], the difference may be explained by cultural and ethnical contrasts.

Studies that have assessed factors associated to these symptoms, considering physical activity as an exposure, are scarce. Therefore, this study aimed at measuring the prevalence of aging symptoms and associated factors among men 40 years or older, living in the urban area of Pelotas, Brazil.

Methods

Pelotas is a Southern Brazil city with nearly 340,000 inhabitants, and almost 32% (112,000) are 40 or older according to the 2000 Census [12]. During 2007, we carried out a cross-sectional observational study among men (40 years or older) living in the urban area of the city.A clustered multistage sampling was employed. First, all urban census tracts were listed (n = 408), based on the IBGE 2000 Census. Four tracts were excluded because they included collective places (such as prisons). Later, the first tract to be included in the sample was randomly selected. Using this tract as a starting point, 45 tracts were chosen on a systematic basis (one out of each nine census tracts). Within each tract a starting point was established and based on that point, 20 houses were chosen for the sample. We excluded men presenting severe motor disabilities (tetraplegic men or cerebral palsy patients) and those unable to respond or to understand the questionnaire (mentally challenged men).

To measure the prevalence of the aging male's symptoms, we administered an adaptation of the The Aging Males' Symptoms Scale (AMS) (Appendix 1). The original instrument was in Portuguese (from Portugal) and was altered to suit the Brazilian Portuguese. The instrument was composed of 17 items (www.aging-males-symptoms-scale.info) [13]. The scale is subdivided in subscales (psychological, somatic and sexual) and the sum of each subscale (1–5 points) results in the total score [10]. Based on this scoring system, two analyses were performed:

• In the first analysis, the main outcome was dichotomous, with a cutoff point of equal or above 37 points. Men above this cutoff point were considered as having moderate/severe symptoms, while those below were classified as not having/mild symptoms [11].

• In the second analysis the outcome was categorized in four groups as follows: 17–26 points (no symptoms); 27–36 points (mild symptoms); 37–49 points (moderate symptoms); and 50+ (severe symptoms).

To classify men as physically active or inactive, the long version of the International Physical Activity Questionnaire (IPAQ) was used. The instrument includes 27 questions about four domains of physical activity (leisure, commuting, occupation, and household). Men attaining at least 150 min/week of physical activities were considered active, according to the ACSM recommendation [14]. Except for the sexual symptoms questionnaire, all instruments were administered during in-person interviews. Sexual symptoms were assessed by a confidential questionnaire to minimize refuse (men were handed an envelope containing the questionnaire and after filling it the envelope was sealed). Those unable to read or to understand the questions could ask for interviewer's help.

The following variables were collected by a pre-tested standardized questionnaire: age (years); skin color (white or non-white, according to interviewer's perception); marital status (living with or without partner); economic level (ABEP 2008 [15] – A; B; C; D/E); schooling (complete years of formal education); smoking (current smoker, non-smoker or former smoker); Body Mass Index (BMI – normal, overweight or obese); and self-perceived health status (excellent, very good, good, fair, poor). The instrument was tested during a pilot study.

The instrument employed included a section on sexual intercourse frequency, libido and morning erections. Such information was collected by a confidential questionnaire. Those unable to read or understand the questions (<10% of the sample) were helped accordingly by the interviewer. The staff worked always in an attempt to keep the answers as confidential as possible. For the sample size calculation, we estimated a 20% prevalence of aging symptoms among men, aged 40 or older; an error margin of 4 percentage points; and 95% significance level. Initially, the estimative resulted in 384 men. Later, 10% was added to compensate refuses, resulting in a sample size of 422 subjects. The questionnaires were administered by a trained staff (high school graduates attending 40 h of training) that was not aware of the intentions of the survey. Fieldwork supervisors interviewed 10% of the sample, randomly selected, using a short version of the questionnaire to perform quality control of the survey.

The results of the study were compared to a German and a Japanese study that have used the same instrument [10,11]. The German study was carried out with a representative sample of the country; while the Japanese study evaluated men that were being followed in a health service. The age groups studied in the three studies are similar but, socioeconomically many differences were observed. The dataset was built on EpiInfo 6.0, data were entered twice. The statistical package STATA 9.0 was used for analysis. We first described subjects with respect to male's aging symptoms, physical activity level and socioeconomic, demographic, behavioral and health variables. Linear trend and chi-square tests were performed during crude analysis. The multivariable analysis, by Poisson regression, followed a four-level hierarchical model. The distal level included age and skin color; the second level included schooling, economic level, and marital status; third level – BMI, smoking, and physical activity; and the self-perceived health status and aging male's symptoms were included in the last level. The effects of the variables were controlled for the same level and upper level covariates. We retained in the model those presenting p-values below 0.2 and the significance level was set to 5%. All subjects signed a consent form before the interview and the study was approved by the Research Ethics Committee of the Physical Education School of the Federal University of Pelotas.

Results

A total of 421 men (40 years or older) were interviewed at 876 houses, losses and refuses totaled 8.3%. The design effect observed (0.7) did not affect the statistical power and confidence level predicted for the study. Mean values for age and schooling were, respectively, 54.5 ± 10.5 and 7.2 ± 3.9 years, 16.7% of men attended <4 years of formal education. Two-thirds were still working regularly. Median time spent on physical activity was 223 min per week.

Nearly 85% of individuals were white, 46.7% belonged to the economic level C and 77.2% were married or lived with a partner. Twenty percent of men were current smokers, 37.2% were physically inactive and 21.8% were obese (Table I). The main outcome of the study – aging male's symptoms (moderate/severe symptoms) – was considered positive (score equal or above 37 points on AMS scale) in 20% of men. The mean score was 29.9 ± 8.4 points, ranging from 17 to 70 points. The scoring distribution (total and subscales) according to age indicates that men older than 70 years were more likely to present severe symptoms (total, psychological and sexual subscales), but somatic symptoms were more in the 60–69 years age group. Men between 40 and 49 years presented less aging symptoms, except for psychological symptoms (Table II).

Table I.  Characteristics of the studied men (40 years or older) in Pelotas – Brazil, 2007 (n = 421).

Variables	n	%
Age (years)
40–49	164	39.0
50–59	137	32.6
60–69	77	18.3
70 or older	43	10.1
Skin color
White	357	85.2
Non white	62	14.8
Schooling (years)
0	14	3.3
1–4	112	26.6
5–8	146	34.7
9–11	76	18.1
12 or more	73	17.3
Economic level
A (highest)	33	8.0
B	140	33.9
C	191	46.2
D/E	49	11.9
Marital status
Living with partner	325	77.2
Living without partner	96	22.8
Body Mass Index
Normal	133	32.9
Overweight	183	45.3
Obesity	88	21.8
Smoking status
Non-smoker	128	30.4
Former smoker	175	41.6
Current smoker	118	28.0
Physical activity level
Inactive	262	62.8
Active	155	37.2
Self-perceived health
Excellent	41	9.8
Very good	58	13.8
Good	216	51.4
Fair	85	20.2
Poor	20	4.8
Moderate/severe symptoms*
No	312	80.0
Yes	78	20.0

*Variable with highest missing count (31).

Table II.  Severity of aging male's symptoms, stratified by age in Pelotas – Brazil, 2007 (n = 421).

		Symptoms classification n (%)
Score	Age (years)	None/little	Mild	Moderate	Severe
Total	≥70	11 (6.5)	12 (8.5)	13 (19.1)	5 (50.0)
	60–69	24 (12.1)	33 (23.2)	14 (20.6)	2 (20.0)
	50–59	56 (32.9)	47 (33.1)	22 (32.4)	1 (10.0)
	40–49	79 (46.5)	50 (35.2)	19 (27.9)	2 (20.0)
Psychological	≥70	10 (23.3)	15 (34.9)	7 (16.3)	11 (25.6)
	60–69	18 (23.4)	31 (40.3)	18 (23.4)	10 (13.0)
	50–59	27 (20.2)	47 (35.1)	37 (27.6)	23 (17.2)
	40–49	16 (9.9)	76 (47.2)	40 (24.8)	29 (18.0)
Somatic	≥70	7 (16.3)	15 (34.9)	16 (37.2)	5 (11.6)
	60–69	13 (16.9)	32 (41.6)	22 (28.6)	10 (13.0)
	50–59	28 (20.9)	63 (47.0)	33 (24.6)	10 (7.5)
	40–49	35 (21.7)	73 (45.3)	46 (28.6)	7 (4.4)
Sexual	≥70	3 (7.3)	2 (4.9)	9 (22.0)	27 (65.9)
	60–69	5 (6.9)	9 (12.3)	26 (35.6)	33 (45.2)
	50–59	14 (11.1)	25 (19.8)	53 (42.1)	34 (27.0)
	40–49	29 (19.3)	52 (34.7)	49 (32.7)	20 (13.3)

The psychological, somatic, and sexual aging symptoms found were compared to the results from Japan [9] and German [8] (Figure 1). In terms of psychological symptoms, we observed a higher prevalence (42.3%) compared to Japan (32.2%) and German (14.6%) (p = 0.01 and p < 0.001, respectively). The somatic symptoms were more frequent among Japanese (49.0%) compared to Brazilian (35.9%) (p = 0.003) and our prevalence was also higher than the German (26.4% and p = 0.04). The frequency of sexual symptoms observed was also higher than in the German population (64.3% vs. 27.8%; p < 0.001). Only 6.4% of subjects did not respond to the sexual questions.

Figure 1.  Prevalence of psychological, somatic and sexual aging symptoms in representative samples of Brazilian, Japanese and German men.

*Significant difference from the Brazilian to the Japanese and German population (p = 0.01; p < 0.001, respectively).

^#Significant difference from the Japanese population to the Brazilian and from the Brazilian to the German population (p = 0.003; p = 0.04, respectively).

^£Significant difference from the Brazilian to German population (p < 0.001).

The crude analysis (Table III) showed that the aging male's symptoms were directly associated to age and inversely associated to schooling, socioeconomic level, and self-perceived health status. Other variables associated were marital status, smoking, and physical activity level. After multivariable analysis (Table III), smoking, age, physical activity level, and self-perceived health status were associated to the outcome.

Table III.  Prevalence of moderate/severe aging male's symptoms (scores ≥37 points) and association with sociodemographic, socioeconomic and behavioral variables in Pelotas – Brazil, 2007 (n = 421).

			Crude analysis		Adjusted analysis
Variables	n	%	PR (CI 95%)	p	PR (CI 95%)	p
Age (years)^†				<0.001**		<0.001**
40–49	21	14.0	1.0		1.0
50–59	23	18.3	1.3 (0.8–2.2)		1.3 (0.8–2.3)
60–69	16	21.9	1.6 (0.9–2.8)		1.6 (0.9–2.9)
70 or older	18	43.9	3.1 (1.9–5.3)		3.1 (1.8–5.2)
Skin color^†				0.1		0.2
White	64	19.2	1.0		1.0
Non-white	14	25.9	1.3 (0.9–1.7)		1.4 (0.8–2.2)
Schooling (years)^‡				0.004**		0.07
0	6	46.2	1.0		1.0
1–4	27	27.0	0.6 (0.3–1.1)		0.8 (0.4–1.5)
5–8	24	17.8	0.4 (0.2–0.8)		0.6 (0.3–1.1)
9–11	12	16.7	0.4 (0.2–0.8)		0.6 (0.3–1.4)
12 or more	9	12.9	0.3 (0.1–0.7)		0.5 (0.2–1.1)
Economic level^‡				0.01**		0.6
A (highest)	1	3.23	1.0		1.0
B	24	18.6	5.8 (0.8–41.1)		4.5 (0.6 to 32.3)
C	39	21.9	6.8 (1.0–47.8)		4.1 (0.5 to 31.0)
D/E	12	26.1	8.1 (1.1–59.2)		4.1 (0.5–33.6)
Marital status^‡				0.03*		0.06
Living with partner	54	17.8	1.0		1.0
Living without partner	24	27.9	1.6 (1.0–2.4)		1.5 (1.0–2.2)
Body Mass Index^§				0.9		0.7
Normal	26	21.1	1.0		1.0
Overweight	29	17.0	0.8 (0.5–1.3)		1.0 (0.6–1.6)
Obesity	17	21.3	1.0 (0.6–1.7)		1.1 (0.6–2.0)
Smoking status^§				0.004*		0.006*
Non-smoker	12	10.3	1.0		1.0
Former smoker	39	23.6	2.3 (1.3–4.2)		1.7 (0.9–3.1)
Current smoker	27	24.8	2.4 (1.3–4.5)		2.3 (1.2–4.3)
Physical activity level^§				0.02*		0.01*
Inactive	37	26.4	1.0		1.0
Active	40	16.2	0.6 (0.4–0.9)		0.6 (0.4–0.9)
Self-perceived health^¶				<0.001**		<0.001**
Excellent	3	8.1	1.0		1.0
Very good	5	9.3	1.1 (0.3–4.5)		1.2 (0.2–6.0)
Good	28	13.7	1.7 (0.5–5.3)		2.1 (0.5–8.6)
Fair	30	40.5	5.0 (1.6–15.3)		5.5 (1.3–22.6)
Poor	12	63.2	7.8 (2.5–24.3)		6.0 (1.3–26.9)

PR, prevalence ratios; CI95%, 95% confidence intervals.

^†Adjusted for age and skin color.

^‡Adjusted for variables in level 1 and schooling, economic level and marital status.

^§Adjusted for variables in levels 1 and 2, and body mass index, smoking and physical activity.

^¶Adjusted for variables in the three levels above.

*Wald's test for heterogeneity.

**Wald's test for trend.

Physical inactivity was associated to the symptoms' severity (p = 0.009), among inactive people, 70% presented severe symptoms (Figure 2). The prevalence of aging male's symptoms increased linearly with age, the 70-year-old age group presented a threefold increase, compared to the age group 40–49. Former and current smokers presented a twofold increase compared to men who never smoked. Physically active men presented a 40% protection against aging male's symptoms compared to inactive peers. Self-reported health status was inversely associated to the aging symptoms, those perceiving their health as poor were six times more likely to present symptoms compared to those with excellent self-perceived health (Table III).

Figure 2.  Severity of aging symptoms according to the physical activity level.

Discussion

The aging male's symptoms were previously studied in other settings [10,11,16] as part of follow-up health studies [11] and also to evaluate the need for hormonal treatment among men presenting symptoms [17,18]. The prevalence of aging male's symptoms in the studied population was 20%, age, smoking; physical inactivity and self-perceived health were associated to the symptoms. Other studies reported similar prevalence, combining moderate and severe symptoms into a single category resulted in 18–22.7% prevalence [10,11].

Comparing our sample with the German study [10], sexual, somatic, and psychological symptoms were different. Similarly, the comparison between the German and the Japanese study also revealed distinct results [11]. However, when our results were compared to the Japanese study, our sample presented more psychological symptoms while among the Japanese somatic symptoms were more prevalent. Before further conclusions, we must consider that the Japanese sample included men attending a health service, and group relationships may positively influence the socialization process, minimizing symptoms or leading men to behavioral changes. Besides, comparisons must be taken cautiously since socioeconomic and educational differences certainly influence the aging process distinctly in each country.

With respect to age, symptoms were aggravated by aging [9–11], in agreement with Nigerian results, which report prevalence rates around 37.0% for the age group 60–79 [16]. The increase is expected, because with age the death risk, chronic conditions, and disabilities increase due to physiological and genetic aging processes [3]. A negative influence of smoking was observed. Non-smokers were less likely to present aging symptoms compared to former and current smokers. Field et al. previously reported that smoking is associated with a small increase in sexual hormones, regardless of body weight and age [19]. However, regardless of this increase, the cardiovascular risk may increase as well, counterbalancing the hormonal alteration resulting from smoking and increasing the risk of aging symptoms. Besides, quality of life is undoubtedly worst, smokers have poorer health perception and life expectancy decreases in nearly 10 years [20]. Other studies have also pointed out aging complications as result of smoking [21,22].

In this study, self-perceived health was associated to the aging male's symptoms, men reporting poorer health status presented more serious symptoms. The self-perceived health status has been identified as a mortality indicator in the US [23] and associated to chronic diseases in men, the perception is poorer among men presenting four or more chronic conditions [24,25]. Physical inactivity was also positively associated to aging symptoms. There is evidence that psychological symptoms are affected positively by physical activity, especially anxiety and depression [26], cognitive function [27]; and sleep disorders are less frequent in active people [28].

With respect to somatic symptoms, physical activity contributes to the conservation of strength and muscle mass, reducing fracture risks and improving quality of life [10,29]. Physical activity can also increase muscle mass and strength [30], flexibility [10], oxidative capacity, muscle fiber capilarization [10,31], and bone mineral density [32], improves blood lipid profile [33] and reduces body fat [34], which are aging somatic characteristics.

Physical activity may affect positively two of the sexual symptoms, erectile dysfunction and low libido. Physical inactivity has been associated to erectile dysfunction [35,36], middle-aged men engaging in physical activities have lower risks of erection problems [37], and weekly calorie expenditure above 1000 Kcal decreases the chance of such condition [38]. Cardiovascular disease can also affect sexual activity by three mechanisms: the psychological impact of the diagnosis may lead to anxiety, fear of death and physical activity restrictions; medicine use may impair sexual activity by decreasing libido and leading to erectile dysfunction; and physiologically the erection function may be affected negatively by vascular conditions. On the other hand, physically active men present lower risks of such conditions; are relatively protected against myocardial infarction; and are more prompted for sexual activity [39]. Therefore, physical activity may act as a preventive factor against the aging male's symptoms.

Chronic diseases represent a large economic burden worldwide [40]. In Brazil, in 2006, the health public expenses achieved nearly 10 billion US dollars, a net increase of 7.3% from 2001 to 2006 [41]. With population aging, the tendency for this budget is to increase, as consequence of associated diseases [42]. Thus, the prevention of such conditions would impact enormously health costs and should be part of the priorities of primary health care services. Overall, we conclude that healthy lifestyles, not smoking and regular physical activity can slow down aging symptoms and increase life expectancy [43], postponing the onset of associated diseases. From a cost-effective point of view, it seems to be more adequate to prevent than treat when dealing with such health issues.

Although the AMS scale is validated in many countries, including Portugal, a Brazilian validation has not been carried out. The validation is desired – although Brazil and Portugal share the same language (Portuguese), the differences between the two countries are enormous and even within Brazil the regional and socioeconomic contrasts indicate the need for regional adaptations. This is the first population-based Brazilian study, and one of the few to assess the aging male's symptoms and associated factors in a sample with such a low refuse rate. Among the limitations of the study, reverse causality is a potential problem inherent of any cross-sectional study (collection of outcomes and risk factors at the same time), especially with respect to physical activity level and self-perceived health status. An information bias could have happened and affected the results. Men could overestimate their sexual behaviors, resulting in a decrease of overall and sexual scoring of symptoms. However, considering the high prevalence observed, it does not seem to have occurred.

Although aging is an irreversible process, the extent to which aging negatively affects life quality may be largely changed by lifestyle characteristics like daily physical activity and not smoking – factors associated to aging symptoms in our study. Public health initiatives aimed at improving these healthy habits would minimize aging symptoms and, therefore, lower the number of people affected by conditions associated to aging. Besides, an absolute understanding of the problem may be only achieved after longitudinal studies are carried out to fully explore the causal pathways.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.