Age-Related Changes in General and Sexual Health in Middle-Aged and Older Men: Results from the European Male Ageing Study (EMAS)jsm_

Transcription

1 1362 ORIGINAL RESEARCH EPIDEMIOLOGY Age-Related Changes in General and Sexual Health in Middle-Aged and Older Men: Results from the European Male Ageing Study (EMAS)jsm_ Giovanni Corona, MD,* David M. Lee, PhD, MPH, Gianni Forti, MD,* Daryl B. O Connor, PhD, Mario Maggi, MD,* Terence W. O Neill, MD, Neil Pendleton, MD, Gyorgy Bartfai, MD, Steven Boonen, MD,** Felipe F. Casanueva, MD, PhD, Joseph D. Finn, BSc, Aleksander Giwercman, MD, PhD, Thang S. Han, MD, Ilpo T. Huhtaniemi, MD, PhD, Krzysztof Kula, MD, PhD,*** Michael E.J. Lean, MD, Margus Punab, MD, Alan J. Silman, MD, Dirk Vanderschueren, MD, PhD, Frederick C.W. Wu, MD, and the EMAS Study Group *Andrology Unit, Department of Clinical Physiopathology, University of Florence, Florence, Italy; arc Epidemiology Unit, The University of Manchester, Manchester, UK; Institute of Psychological Sciences, University of Leeds, Leeds, UK; Clinical Gerontology, The University of Manchester, Hope Hospital, Salford, UK; Department of Obstetrics, Gynaecology and Andrology, Albert Szent-Gyorgy Medical University, Szeged, Hungary; **Division of Gerontology and Geriatrics & Centre for Musculoskeletal Research, Department of Experimental Medicine, Katholieke Universiteit Leuven, Leuven, Belgium; Department of Medicine, Santiago de Compostela University, Complejo Hospitalario Universitario de Santiago (CHUS); CIBER de Fisiopatología Obesidad y Nutricion (CB06/03), Instituto Salud Carlos III; Santiago de Compostela, Spain; Reproductive Medicine Centre, Malmö University Hospital, University of Lund, Sweden; Department of Endocrinology, Royal Free and University College Hospital Medical School, Royal Free Hospital, Hampstead, London, UK; Department of Reproductive Biology, Imperial College London, Hammersmith Campus, London, UK; ***Department of Andrology and Reproductive Endocrinology, Medical University of Lodz, Lodz, Poland; Department of Human Nutrition, University of Glasgow, Glasgow, UK; Andrology Unit, United Laboratories of Tartu University Clinics, Tartu, Estonia; Department of Andrology and Endocrinology, Katholieke Universiteit Leuven, Leuven, Belgium; Department of Endocrinology, Manchester Royal Infirmary, The University of Manchester, Manchester, UK DOI: /j x ABSTRACT Introduction. Limited information is available concerning the general and sexual health status of European men. Aim. To investigate the age-related changes in general and sexual health in middle-aged and older men from different countries of the European Union. Methods. This is a cross-sectional multicenter survey performed on a sample of 3,369 community-dwelling men aged years old (mean years). Subjects were randomly selected from eight European centers including centers from nontransitional (Florence [Italy], Leuven [Belgium], Malmö [Sweden], Manchester [United Kingdom], Santiago de Compostela [Spain]) and transitional countries (Lodz [Poland], Szeged [Hungary], Tartu [Estonia]). Main Outcome Measures. Different parameters were evaluated including the Beck s Depression Inventory for the quantification of depressive symptoms, the Short Form-36 Health Survey for the assessment of the quality of life (QoL), the International Prostate Symptom Score for the evaluation of lower urinary tract symptoms, and the European Male Ageing Study sexual function questionnaire for the study of sexual function. Results. More than 50% of subjects reported the presence of one or more common morbidities. Overall, hypertension (29%), obesity (24%), and heart diseases (16%) were the most prevalent conditions. Around 30% of men reported erectile dysfunction (ED) and 6% reported severe orgasmic impairment, both of which were closely associated with age and concomitant morbidities. Only 38% of men reporting ED were concerned about it. Furthermore, concern about The EMAS Study Group: Florence (Gianni Forti, Luisa Petrone, Antonio Cilotti); Leuven (Dirk Vanderschueren, Steven Boonen, Herman Borghs); Lodz (Krzysztof Kula, Jolanta Slowikowska-Hilczer, Renata Walczak-Jedrzejowska); London (Ilpo Huhtaniemi); Malmö (Aleksander Giwercman); Manchester (Frederick Wu, Alan Silman, Neil Pendleton, Terence O Neill, Joseph Finn, Philip Steer, Abdelouahid Tajar, David Lee, Stephen Pye); Santiago (Felipe Casanueva, Mary Lage); Szeged (Gyorgy Bartfai, Imre Földesi, Imre Fejes); Tartu (Margus Punab, Paul Korrovitz); Turku (Min Jiang) 2009 International Society for Sexual Medicine

2 General and Sexual Health in European Men 1363 ED increased with age, peaking in the years age band, but decreased thereafter. Men in transitional countries reported a higher prevalence of morbidities and impairment of sexual function as well as a lower QoL. Conclusion. Sexual health declined while concomitant morbidities increased in European men as a function of age. The burden of general and sexual health is higher in transitional countries, emphasizing the need to develop more effective strategies to promote healthy aging for men in these countries. Corona G, Lee DM, Forti G, O Connor DB, Maggi M, O Neill TW, Pendleton N, Bartfai G, Boonen S, Casanueva FF, Finn JD, Giwercman A, Han TS, Huhtaniemi IT, Kula K, Lean MEJ, Punab M, Silman AJ, Vanderschueren D, Wu FCW, and EMAS Study Group. Age-related changes in general and sexual health in middle-aged and older men: Results from the European Male Ageing Study (EMAS).. Key Words. Aging; Male Health; Population Survey; Sexual Function; Quality of Life; Sexual Health Introduction In past decades, there has been a disproportionate focus on women s health, while men s health has not received similar attention. In all industrialized countries, and especially in the transitional countries of eastern Europe, men have lower life expectancy than women [1 3]. Life expectancy has been increasing in western European countries since 1950; however, for men living in eastern European (transitional) countries, life expectancy has not only remained unchanged, but a further decrease (currently about 6 7 years lower) [1 5] is predicted for 2020 [4]. Accordingly, the Disease Control Priorities Project [6] demonstrated that death rates from cardiovascular diseases (CVD) among adults aged under 60 years between 1990 and 2001 have fallen worldwide except in the low- and middle-income eastern countries of Europe and Central Asia. There is a consensus that changes in health are related to the deterioration in social and economic conditions in all European transitional countries [2,5,7,8]. Erectile dysfunction (ED) has been proposed as an indicator of male health [9 11]. Although 10% to 20% of men in the general population of many countries are said to be affected, data on the prevalence of ED in transitional countries are lacking [12]. The European Male Ageing Study (EMAS) is a multicenter, prospective study of aging in eight European centers, including three from eastern Europe. Aim The aim of this study is to investigate the agerelated changes in general and sexual health in middle-aged and older men from different countries of the European Union. Methods Subjects and Study Design Three thousand three hundred sixty-nine men aged years (mean age years) were recruited from population registers in eight European centers (Florence [Italy], Leuven [Belgium], Lodz [Poland], Malmö [Sweden], Manchester [United Kingdom], Santiago de Compostela [Spain], Szeged [Hungary], Tartu [Estonia]). Ethical approval for the study was obtained in each of the centers in accordance with local practice and requirements. The term transitional country is often used to cover the countries of central and eastern Europe and the Former Soviet Union ( wiki/transition_economy); hence, this term will be used in the rest of the article to indicate men living in Lodz, Szeged, and Tartu. The choice of the sampling frame was limited by the availability of specific registers within each center. Registers used included general practitioner age sex registers (Florence, Manchester, Tartu), population or national registers (Malmö, Santiago), electoral registers (Leuven, Szeged), and a city register (Lodz). The population registers were, to the knowledge of the participating centers, those most current at the onset of the study. There were no specific exclusion criteria apart from subjects being able to provide written, informed consent. Stratified random sampling was used with the aim of recruiting equal numbers of men into each of the four age bands (40 49, 50 59, and years). Subjects were invited with a letter of invitation to attend a screening at a local clinic as previously described [13]. For those patients who had not replied after two attempts, a sample of them were contacted by telephone, inviting them to verbally answer a series of questions taken from the postal questionnaires ([13], see below).

3 1364 Corona et al. Subjects initially completed a postal questionnaire, which included questions about selfreported health (excellent, very good, good, fair, poor), employment (paid or voluntary, none), age on leaving full-time education (years), tobacco use (nonsmoker, current smoker), and frequency of alcohol consumption (none, <once/week, 1 2 days/ week, 3 4 days/week, 5 6 days/week, every day). The postal questionnaire also included a range of questions about comorbidities, Are you currently being treated for any of the following conditions?, including: heart conditions, high blood pressure, prostate diseases, and diabetes. The occurrence of cerebrovascular disease was assessed using the question, Have you ever been told by a doctor that you have had a stroke?, and cancer from the question, Have you ever been diagnosed as having cancer? After completing the postal questionnaire, subjects attended research clinics for a health screening. Each participant completed interviewer-assisted questionnaires and underwent clinical assessments. The questionnaires included Beck s Depression Inventory (BDI), the Short Form-36 Health Survey (SF-36), the International Prostate Symptom Score for lower urinary tract symptoms (LUTS), medication use, and previous surgical procedures [13]. Subjects then completed the EMAS sexual function questionnaire (EMAS-SFQ; 14, see appendix A) in private. Height and weight were also measured [13]. Statistics Descriptive statistics were used to summarize the response variables from the EMAS-SFQ by age decade and center. One-way analysis of variance (anova) or the chi-square test was used to compare values or proportions between groups where applicable. When overall nonresponse for a particular question exceeded 2%, the nonresponders were removed from the denominator to allow clear comparisons to be made between groups (decades/centers). Multivariate logistic regression analysis with adjustments for age and comorbidities was used to model the likelihood of experiencing specific sexual problems and of being in the lowest quartile of the physical and mental components of SF-36. All analyses were performed using SPSS version 14 (SPSS Inc., Chicago, IL, USA), and Intercooled STATA version 9.2 (StataCorp, College Station, TX, USA). Results Subjects Of 8,416 men invited, 3,369 (mean age years) participated in the study (response rate 41%). Comparing subjects who participated in the study with those who did not, no differences in the number of morbidities, current and past smoking, and the frequency of alcohol consumption were found (not shown). Complete data concerning frequency of sexual activities (excluding men without a sexual partner) and erectile function were available in 2,734 (92%) and 3,193 (95%) of subjects, respectively, without significant differences between centers (data not shown). Table 1 summarizes the demographics of the EMAS sample. There were education differences (age on leaving full-time education) between centers (P < 0.05, one-way anova), but not between age decades. About 85% reported living with their wife/partner, while the proportion of single and employed men was higher and lower, respectively, in the higher age bands. The highest proportion of single men was observed in Malmö and Tartu. Those employed full time (paid or voluntary) ranged from 49% in Lodz to 85% in Szeged. When subjects living in transitional countries were compared with those in nontransitional countries, no significant differences were observed in marital status, although subjects in transitional centers were more often employed and left fulltime education later in life. General Health Table 2 presents physical and self-reported health status. The health characteristics differed between the four decades (P < 0.05, c 2 test) except for obesity (body mass index [BMI] 30 kg/m 2 ) and use of antidepressants. The proportion of men reporting their own or their partner s health as fair/poor and the prevalence of chronic diseases were higher in the older age groups. Health profiles differed significantly between centers (P < 0.05, c 2 test), except for cancer. Tartu and Szeged had the highest prevalence of obesity, Florence the lowest. The proportion of men reporting fair/poor personal or partner health was highest in Tartu and lowest in Manchester. Szeged reported the highest prevalence of stroke, hypertension, and diabetes, and highest use of antihypertensive and antidiabetic medications. The prevalence of LUTS, nocturia, and treatment for prostate diseases was higher in the older age groups. The occurrence of moderate-to-severe

4 General and Sexual Health in European Men 1365 Table 1 Demographic characteristics Age band (years) (n = 796) (n = 904) (n = 839) 70 and over (n = 830) All (n = 3,369) Mean (SD) Age left education (years) 20.8 (5.8) 20.7 (6.5) 20.8 (8.1) 21.1 (9.7) 20.9 (7.7) Marital status Living with wife 585 (75) 703 (80) 674 (83) 602 (77) 2,564 (79) Living with partner 85 (11) 57 (6) 37 (5) 31 (4) 210 (6) Partner, living apart 65 (8) 55 (6) 46 (6) 41 (5) 207 (6) No sexual partner 46 (6) 65 (7) 58 (7) 109 (14) 278 (9) Did not answer question Currently employed* 720 (91) 718 (79) 460 (55) 270 (33) 2,168 (64) Center Florence (n = 433) Leuven (n = 451) Malmö (n = 409) Manchester (n = 396) Santiago (n = 406) Łódź (n = 408) Szeged (n = 431) Tartu (n = 435) Non-Trans. (n = 2,095) Mean (SD) Age left education (years) 16.4 (5.8) 20.2 (4.8) 21.8 (9.0) 17.6 (3.5) 18.1 (6.1) 23.6 (8.3) 25.2 (8.9) 23.9 (7.8) 18.8 (6.4) 24.2 (8.4) Marital status Living with wife 371 (86) 366 (85) 251 (65) 324 (85) 327 (83) 322 (82) 310 (75) 293 (70) 1,639 (81) 925 (75) Living with partner 15 (3) 23 (5) 48 (12) 14 (4) 15 (4) 19 (5) 25 (6) 51 (12) 115 (6) 95 (8) Partner, living apart 25 (6) 17 (4) 40 (10) 15 (4) 15 (4) 24 (6) 41 (10) 30 (7) 112 (6) 95 (8) No sexual partner 21 (5) 27 (6) 50 (13) 30 (8) 36 (9) 28 (7) 40 (9) 46 (11) 164 (8) 114 (9) Did not answer question Currently employed* 236 (55) 248 (55) 252 (62) 254 (64) 260 (64) 199 (49) 368 (85) 251 (81) 1,250 (60) 818 (72) Trans. (n = 1,274) *Includes paid and/or voluntary work. Between age band or between center differences P < 0.05 except. Due to rounding, percentages may not add up to 100. Non-Trans. = nontransitional centers; SD = standard deviation; Trans. = transitional centers.

5 1366 Corona et al. Table 2 Physical and self-reported health characteristics Age band (years) (n = 796) (n = 904) (n = 839) 70 and over (n = 830) All (n = 3,369) BMI 25 and <30 kg/m (45) 432 (49) 410 (49) 432 (54) 1,629 (49) BMI 30 kg/m (22) 228 (25) 221 (27) 192 (23) 818 (24) General health fair or poor 144 (18) 284 (31) 319 (38) 370 (45) 1,117 (33) Partner s health fair or poor* 87 (12) 159 (21) 238 (35) 268 (44) 752 (27) Did not answer question Morbidities Heart condition 22 (3) 93 (10) 181 (22) 255 (31) 551 (16) High blood pressure 81 (10) 218 (24) 302 (36) 359 (43) 960 (29) Stroke 7 (1) 19 (2) 33 (4) 67 (8) 126 (4) Diabetes 17 (2) 47 (5) 82 (10) 108 (13) 254 (8) Cancer 13 (2) 29 (3) 67 (8) 90 (11) 199 (6) Prostate disease 10 (1) 47 (5) 125 (15) 222 (27) 404 (12) No reported morbidities 620 (78) 517 (57) 306 (36) 210 (25) 1,588 (47) Medication use Hypertension 71 (9) 206 (23) 310 (37) 388 (47) 975 (29) Lipid lowering 23 (3) 87 (10) 153 (18) 158 (19) 421 (13) Diabetic 10 (1) 38 (4) 58 (7) 80 (10) 186 (6) Depression 19 (2) 35 (4) 36 (4) 34 (4) 124 (4) Prostate 3 (0) 19 (2) 64 (8) 87 (10) 173 (5) IPSS category Moderate 94 (12) 155 (18) 233 (29) 245 (31) 727 (22) Severe 6 (1) 30 (3) 57 (7) 65 (8) 158 (5) Did not answer question Nocturia 1 2 times/night 370 (47) 485 (55) 518 (63) 561 (71) 1,934 (59) 3 times/night 75 (10) 143 (16) 159 (19) 166 (21) 543 (16) Did not answer question Surgery Prostate 1 (0) 5 (1) 31 (4) 108 (13) 145 (4) Genitourinary 53 (7) 65 (7) 64 (8) 53 (6) 235 (7) Center Florence (n = 433) Leuven (n = 451) Malmö (n = 409) Manchester (n = 396) Santiago (n = 406) Łódź (n = 408) Szeged (n = 431) Tartu (n = 435) Non-Trans. (n = 2,095) Trans. (n = 1,274) BMI 25 and <30 kg/m (52) 207 (48) 184 (47) 215 (55) 215 (53) 221 (54) 187 (44) 176 (42) 1,045 (51) 584 (47) BMI 30 kg/m 2 74 (17) 86 (19) 84 (21) 84 (21) 110 (27) 89 (22) 154 (36) 137 (32) 438 (21) 380 (30) General health fair or poor 140 (32) 52 (12) 88 (22) 42 (11) 119 (29) 181 (44) 190 (44) 305 (70) 441 (21) 676 (53) Partner s health fair or poor* 97 (26) 58 (16) 41 (14) 41 (13) 88 (25) 130 (40) 106 (32) 191 (49) 325 (19) 427 (40) Did not answer question Morbidities Heart condition 51 (12) 71 (16) 39 (10) 45 (11) 56 (14) 116 (28) 74 (17) 99 (23) 262 (13) 289 (23) High blood pressure 125 (29) 125 (28) 94 (23) 82 (21) 99 (24) 131 (32) 164 (38) 140 (32) 525 (25) 435 (35) Stroke 7 (2) 15 (3) 19 (5) 15 (4) 16 (4) 13 (3) 29 (7) 12 (3) 72 (3) 54 (4) Diabetes 17 (4) 26 (6) 21 (5) 29 (7) 45 (11) 43 (11) 55 (13) 18 (4) 138 (7) 116 (9) Cancer 23 (5) 22 (5) 34 (8) 31 (8) 19 (5) 17 (4) 25 (6) 28 (6) 129 (6) 70 (6) Prostate disease 64 (15) 50 (11) 25 (6) 30 (8) 54 (13) 77 (19) 65 (15) 39 (9) 223 (11) 181 (15) No reported morbidities 206 (48) 232 (51) 234 (57) 223 (56) 182 (45) 132 (32) 167 (39) 211 (49) 1,077 (51) 510 (40) Medication use Hypertension 124 (29) 160 (35) 89 (22) 89 (22) 121 (30) 101 (25) 160 (37) 131 (30) 583 (28) 392 (31) Lipid lowering 44 (10) 74 (16) 54 (13) 68 (17) 80 (20) 35 (9) 52 (12) 14 (3) 320 (15) 101 (8) Diabetic 13 (3) 12 (3) 15 (4) 22 (6) 31 (8) 37 (9) 42 (10) 14 (3) 93 (4) 93 (7) Depression 26 (6) 25 (6) 14 (3) 15 (4) 18 (4) 9 (2) 10 (2) 7 (2) 98 (5) 26 (2) Prostate 35 (8) 15 (3) 10 (2) 19 (5) 38 (9) 33 (8) 6 (1) 17 (4) 117 (6) 56 (4) IPSS category Moderate 64 (15) 120 (28) 74 (19) 96 (25) 82 (21) 91 (24) 102 (25) 98 (24) 436 (22) 291 (24) Severe 9 (2) 36 (8) 16 (4) 18 (5) 13 (3) 21 (6) 18 (4) 27 (7) 92 (5) 66 (6) Did not answer question Nocturia 1 2 times/night 254 (59) 256 (58) 259 (66) 254 (64) 230 (57) 187 (47) 258 (61) 236 (57) 1,253 (61) 681 (55) 3 times/night 46 (11) 93 (21) 20 (5) 64 (16) 51 (13) 97 (24) 72 (17) 100 (24) 274 (13) 269 (22) Did not answer question Surgery Prostate 19 (4) 21 (5) 15 (4) 16 (4) 15 (4) 11 (3) 24 (6) 24 (6) 86 (4) 59 (5) Genitourinary 4 (1) 70 (16) 32 (8) 10 (3) 37 (9) 6 (2) 24 (6) 52 (12) 153 (7) 82 (7) *Among those having a sexual partner (N = 2,981). Includes prostatectomy, resection and cancer. Includes vasectomy, phimosis, bladder, and urethral surgery. Between age band or between center differences P < 0.05 except. BMI = body mass index; IPSS = International Prostate Symptom Score. Due to rounding, percentages may not add up to 100. Non-Trans. = nontransitional centers; Trans. = transitional centers.

6 General and Sexual Health in European Men 1367 LUTS differed between centers (P < 0.05, by c 2 test), ranging from 17% in Florence to 36% in Leuven. Overall, men in transitional centers reported poorer health profiles for themselves and their partners as compared with men in nontransitional centers. Accordingly, the prevalence of concomitant morbidities, except for cancer, was higher in transitional centers. No difference in the prevalence of moderate-to-severe LUTS was observed between transitional and nontransitional centers; however, the prevalence of subjects reporting nocturia more than tree times per night was significantly higher in transitional centers (P < 0.05, by c 2 test). Lifestyle and Quality of Life (QoL) Measures Both smoking and alcohol consumption showed an inverse relationship with age (Table 3). Smoking was highest in Tartu and lowest in Manchester, while the frequency of alcohol consumption was highest in Manchester and lowest in Lodz. Depressive symptoms (BDI) were most prevalent in Lodz and least prevalent in Malmö in agreement with the center distribution of the SF-36 mental component score (Table 3). Taken together, men from transitional centers were more likely to smoke and consume alcohol less frequently than men from nontransitional centers. In addition, higher levels of depressive symptomology, together with poorer QoL (lower SF-36 mental and physical component scores), were observed in the transitional centers (Table 3). Frequency of Sexual Activity Table 4 reports the frequency of sexual activities. Frequency of sexual intercourse, kissing and petting, and masturbation was lower in the older age groups. Sixteen percent of men reported no sexual intercourse and almost 59% reported they had not masturbated in the preceding 4 weeks. In comparison with the oldest age group ( 70 years), the proportion of men concerned about sexual activity was higher in the and year age groups. Florence reported the highest frequency of sexual intercourse (92% of subjects reported at least one attempt in the previous 4 weeks), while Tartu (73%) and Manchester (76%) reported the lowest among transitional and nontransitional countries, respectively. Tartu reported the lowest frequency of sexual thoughts, petting, and masturbation. Conversely, Lodz, and in particular Szeged, reported similar prevalence to that observed in nontransitional centers for different sexual activities including thinking about sex, sexual intercourse, and petting. Concern about frequency of sexual activity ranged from 8% in Santiago to 16% in Leuven. Overall, no differences were seen between transitional and nontransitional centers for frequency of sexual thoughts, petting, and intercourse. Similar results were observed for concern about frequency of sexual activities. Conversely, subjects living in nontransitional centers reported a higher frequency of masturbation. Prevalence of Erectile and Orgasmic Dysfunction ED (moderate or severe) was reported in 30% of the entire EMAS sample (Table 5). The prevalence of ED was higher in the older age groups, peaking in men 70 years and older (64%). Among men with ED, concern about ED was highest (57%) and lowest (28%) in the and 70-years age bands, respectively. There were significant differences in the prevalence of ED (P < 0.05, c 2 test) ranging from 43% in Tartu to 25% or less in Florence, Malmo, Santiago, and Leuven (Table 5); however, the pattern was quite different for reported concern about ED, with 55% in Florence concerned as opposed to 24% in Santiago and 25% in Tartu. Tartu reported the lowest frequency of orgasms (as assessed by achieving orgasm half the time) and the lowest satisfaction regarding the timing of orgasm (proportion who were extremely/highly satisfied; Table 5). Just over half of subjects were satisfied with their sexual relationship, while 83% were satisfied with their nonsexual relationship (Table 5). Satisfaction with nonsexual relationships was independent of age, while the proportion of men satisfied with their sexual relationship was lower in the oldest age decade (43%) compared with the youngest (60%). While most men were satisfied with their nonsexual relationships (71% in Szeged to 87% in Malmo, Table 5), there was greater variability in satisfaction with sexual relationships between centers: 42% in Szeged to 64% in Santiago satisfied and 35% in Manchester and 20% in Tartu dissatisfied (Table 5). Although men living in transitional centers reported a higher frequency of ED, they were less concerned about it as compared with men in nontransitional centers. In addition, men in transitional centers reported higher levels of orgasm difficulties (66% vs. 75% reporting they achieved orgasm half the time) and lower satisfaction in

7 1368 Corona et al. Table 3 Lifestyle, QoL, and physical performance characteristics Age band (years) (n = 796) (n = 904) (n = 839) 70 and over (n = 830) All (n = 3,369) Current smoker* 236 (30) 240 (27) 143 (17) 86 (10) 705 (21) Alcohol intake 1 day/week 473 (59) 532 (59) 470 (56) 400 (48) 1,875 (56) BDI depression category Mild-borderline 115 (14) 161 (18) 145 (18) 180 (22) 601 (18) Moderate 26 (4) 44 (5) 32 (3) 40 (5) 142 (4) Mean (SD) SF-36 quality of life Mental component score 51.0 (8.7) 50.9 (9.7) 52.7 (9.0) 51.6 (10.0) 51.5 (9.4) Physical component score 53.3 (6.1) 50.8 (8.2) 49.0 (8.2) 46.8 (8.7) 50.0 (8.2) Center Florence (n = 433) Leuven (n = 451) Malmö (n = 409) Manchester (n = 396) Santiago (n = 406) Łódź (n = 408) Szeged (n = 431) Tartu (n = 435) Non-Trans. (n = 2,095) Trans. (n = 1,274) Current smoker* 100 (23) 78 (17) 69 (17) 44 (11) 92 (23) 105 (26) 90 (21) 127 (29) 383 (18.5) 322 (25) Alcohol intake 1 day/week 234 (54) 332 (74) 263 (64) 303 (77) 275 (68) 94 (23) 243 (56) 131 (30) 1,407 (67) 468 (37) BDI depression category Mild-borderline 67 (16) 62 (14) 36 (9) 58 (15) 66 (16) 123 (30) 79 (18) 110 (25) 289 (14) 312 (25) Moderate 8 (2) 15 (3) 9 (2) 9 (2) 18 (4) 34 (8) 14 (3) 35 (8) 59 (3) 83 (7) Mean (SD) SF-36 quality of life Mental component score 51.7 (8.3) 51.5 (8.7) 55.0 (9.0) 52.6 (8.6) 54.1 (8.7) 45.3 (9.6) 52.5 (9.8) 49.7 (9.2) 52.9 (8.8) 49.2 (9.9) Physical component score 52.6 (6.1) 50.0 (8.0) 51.8 (7.7) 51.2 (7.6) 51.3 (6.9) 47.8 (8.3) 49.7 (8.4) 45.4 (9.8) 51.4 (7.3) 47.6 (9.1) *Defined as subjects who have smoked at least 100 cigarettes in their entire life and currently smoke cigarettes. Beck s depression inventory: score range for mild borderline = 11 to 16, moderate = 17 and over. Medical Outcomes Study SF-36: mental and physical component summary scores derived from the eight SF-36 subscales using standard scoring algorithm (lower score means lower quality of life). Due to rounding, percentages may not add up to 100. Between age band or between center differences all P < BDI = Beck s Depression Inventory; Non-Trans. = non-transitional centers; QoL = quality of life; SD = standard deviation; SF-36 = Short Form-36 Health Survey; Trans. = transitional centers.

8 General and Sexual Health in European Men 1369 Table 4 Sexual desire and frequency of sexual activities Age band (years) (n = 796) (n = 904) (n = 839) 70 and over (n = 830) All (n = 3,369) Thinking about sex 1 once/week 718 (92) 724 (82) 541 (66) 367 (47) 2,350 (72) <1 once/week 59 (8) 137 (16) 206 (25) 255 (33) 657 (20) Never 5 (1) 17 (2) 67 (8) 155 (20) 244 (8) Did not answer question Frequency of sexual intercourse* 1 once/week 562 (81) 514 (70) 338 (51) 139 (26) 1,553 (59) <1 once/week 109 (16) 176 (24) 206 (31) 187 (35) 678 (26) None 26 (4) 49 (7) 124 (19) 213 (40) 412 (16) Did not answer question Frequency of kissing, petting, etc.* 1 once/week 577 (83) 558 (75) 400 (60) 245 (45) 1,780 (67) <1 once/week 71 (10) 121 (16) 140 (21) 147 (27) 479 (18) None 49 (7) 61 (8) 130 (19) 147 (27) 387 (15) Did not answer question Frequency of masturbation 1 once/week 225 (29) 179 (20) 99 (12) 56 (7) 559 (17) <1 once/week 218 (28) 223 (25) 190 (24) 139 (18) 770 (24) None 338 (43) 474 (54) 519 (64) 569 (74) 1,900 (59) Did not answer question Concerned about frequency of sexual activities 58 (7) 116 (13) 116 (14) 90 (11) 380 (11) Centre Florence (n = 433) Leuven (n = 451) Malmö (n = 409) Manchester (n = 396) Santiago (n = 406) Łódź (n = 408) Szeged (n = 431) Tartu (n = 435) Non-Trans. (n = 2,095) Trans. (n = 1,274) Thinking about sex 1 once/week 342 (79) 336 (77) 193 (50) 319 (83) 325 (83) 255 (65) 322 (78) 258 (62) 1,515 (74) 835 (68) <1 once/week 85 (20) 81 (19) 160 (41) 44 (11) 50 (13) 82 (21) 62 (15) 93 (22) 420 (21) 237 (19) Never 5 (1) 17 (4) 33 (9) 21 (5) 15 (4) 56 (14) 29 (7) 68 (16) 91 (5) 153 (12) Did not answer question Frequency of sexual intercourse* 1 once/week 225 (62) 205 (60) 149 (53) 151 (50) 231 (67) 186 (61) 232 (70) 174 (47) 961 (58) 592 (59) <1 once/week 110 (30) 96 (28) 87 (31) 77 (26) 71 (21) 76 (25) 66 (20) 95 (26) 441 (27) 237 (24) None 27 (7) 43 (13) 47 (17) 73 (24) 41 (12) 45 (15) 35 (11) 101 (27) 231 (15) 181 (18) Did not answer question Frequency of kissing, petting etc* 1 once/week 227 (62) 241 (70) 169 (60) 207 (69) 271 (79) 181 (60) 263 (79) 221 (60) 1,115 (68) 665 (66) <1 once/week 77 (21) 68 (20) 65 (23) 51 (17) 32 (9) 68 (22) 37 (11) 81 (22) 293 (18) 186 (18) None 65 (18) 35 (10) 50 (18) 43 (14) 40 (12) 55 (18) 32 (10) 67 (18) 233 (14) 154 (15) Did not answer question Frequency of masturbation 1 once/week 33 (8) 144 (34) 116 (30) 121 (31) 32 (8) 36 (9) 48 (12) 29 (7) 446 (22) 113 (9) <1 once/week 93 (22) 136 (32) 110 (28) 136 (35) 78 (20) 88 (22) 76 (19) 53 (13) 553 (27) 217 (18) None 296 (70) 147 (34) 163 (42) 130 (34) 276 (72) 268 (68) 286 (70) 334 (80) 1,012 (50) 888 (73) Did not answer question Concerned about frequency of sexual activities 60 (14) 71 (16) 45 (11) 39 (10) 31 (8) 47 (12) 41 (10) 46 (11) 246 (12) 134 (11) *Among those reporting they had a sexual partner in the last 4 weeks (N = 2,981). Between age band or between center differences P < 0.05 except. Due to rounding percentages may not add up to 100. Non-Trans. = non-transitional centers, Trans. = transitional centers.

9 1370 Corona et al. Table 5 Erectile problems, orgasm ability, and satisfaction with sexual and nonsexual relationship Age band (years) (n = 796) (n = 904) (n = 839) 70 and over (n = 830) All (n = 3,369) Erectile dysfunction (ED) Moderate 36 (5) 127 (14) 180 (23) 210 (29) 553 (17) Severe 5 (1) 43 (5) 121 (15) 251 (35) 420 (13) Did not answer question Concerned about erectile ability All respondents 41 (6) 131 (15) 156 (19) 141 (17) 469 (14) Moderate/severe ED (N = 973) 17 (42) 97 (57) 127 (42) 127 (28) 368 (38) Frequency of orgasm Half the time 698 (90) 709 (82) 530 (67) 332 (44) 2,269 (71) <Half the time 28 (4) 36 (4) 57 (7) 56 (7) 177 (6) Rarely/never 23 (3) 41 (5) 61 (8) 84 (11) 209 (7) No sexual activity 23 (3) 80 (9) 141 (18) 278 (37) 522 (16) Did not answer question Satisfaction with timing of orgasm Extremely/highly satisfied 389 (51) 370 (45) 253 (36) 173 (33) 1,184 (42) Moderately/slightly satisfied 346 (45) 415 (50) 385 (55) 268 (51) 1,414 (50) Dissatisfied 34 (4) 45 (5) 68 (10) 82 (16) 229 (8) Did not answer question Satisfaction with overall sex life Satisfied 472 (60) 500 (57) 406 (51) 323 (43) 1,701 (53) Neutral 125 (16) 149 (17) 155 (19) 191 (26) 620 (19) Dissatisfied 185 (24) 226 (26) 235 (30) 228 (30) 874 (27) Did not answer question Satisfaction with non-sexual relationship* Satisfied 562 (82) 608 (83) 530 (81) 451 (81) 2,151 (82) Neutral 55 (8) 49 (7) 51 (8) 51 (9) 206 (8) Dissatisfied 70 (10) 78 (11) 73 (11) 55 (10) 276 (10) Did not answer question Center Florence (n = 433) Leuven (n = 451) Malmö (n = 409) Manchester (n = 396) Santiago (n = 406) Łódź (n = 408) Szeged (n = 431) Tartu (n = 435) Non-Trans. (n = 2,095) Trans. (n = 1,274) Erectile dysfunction (ED) Moderate 60 (14) 80 (19) 41 (11) 58 (15) 54 (14) 86 (23) 72 (18) 102 (25) 293 (15) 260 (22) Severe 45 (11) 55 (13) 50 (13) 64 (16) 36 (9) 47 (13) 48 (12) 75 (18) 250 (12) 170 (14) Did not answer question Concerned about erectile ability All respondents 72 (17) 81 (18) 50 (12) 53 (13) 34 (8) 67 (16) 49 (11) 63 (15) 290 (14) 179 (14) Moderate/severe ED (N = 973) 58 (55) 67 (50) 39 (43) 45 (37) 22 (24) 50 (38) 42 (35) 45 (25) 231 (42) 137 (33) Frequency of orgasm Half the time 320 (76) 294 (70) 292 (76) 296 (77) 280 (75) 240 (64) 307 (76) 240 (58) 1,482 (75) 787 (66) <Half the time 19 (5) 36 (9) 19 (5) 25 (7) 13 (4) 29 (8) 15 (4) 21 (5) 112 (6) 65 (6) Rarely/never 38 (9) 49 (12) 18 (5) 22 (6) 15 (4) 42 (11) 9 (2) 16 (4) 142 (7) 67 (6) No sexual activity 43 (10) 41 (10) 57 (15) 41 (11) 63 (17) 66 (18) 75 (18) 136 (33) 245 (13) 277 (23) Did not answer question Satisfaction with timing of orgasm Extremely/highly satisfied 189 (48) 191 (49) 155 (44) 131 (37) 184 (55) 112 (34) 121 (34) 101 (31) 850 (47) 334 (33) Moderately/slightly satisfied 174 (44) 169 (43) 180 (51) 180 (51) 135 (40) 189 (58) 197 (56) 190 (59) 838 (46) 576 (58) Dissatisfied 31 (8) 32 (8) 18 (5) 39 (11) 17 (5) 24 (7) 35 (10) 33 (10) 137 (7) 92 (9) Did not answer question Satisfaction with overall sex life Satisfied 246 (57) 210 (50) 209 (55) 183 (48) 249 (64) 184 (48) 171 (42) 249 (61) 1,097 (55) 604 (50) Neutral 48 (11) 90 (22) 83 (22) 67 (18) 46 (12) 103 (27) 103 (25) 80 (20) 334 (17) 286 (24) Dissatisfied 134 (31) 118 (28) 90 (24) 132 (35) 95 (24) 94 (25) 130 (32) 81 (20) 569 (28) 305 (26) Did not answer question Satisfaction with nonsexual relationship* Satisfied 319 (85) 285 (82) 245 (87) 244 (84) 277 (81) 247 (81) 225 (71) 309 (83 1,370 (84) 781 (78) Neutral 22 (6) 20 (6) 16 (6) 18 (6) 19 (6) 33 (11) 36 (11) 42 (11) 95 (6) 111 (11) Dissatisfied 33 (9) 42 (12) 20 (7) 30 (10) 47 (14) 26 (8) 58 (18) 20 (5) 172 (10) 104 (10) Did not answer question Erectile difficulties were assessed by a single question from the Massachusetts Male Ageing Study [25]: You are (Always able.../usually able.../sometimes able.../never able...) to get and keep an erection which would be good enough for sexual intercourse. Erectile dysfunction (ED) was coded from this question as No ED (Always able), Mild ED (Usually able), Moderate ED (Sometimes able), and Severe ED (Never able), respectively. We consider only men in the Moderate ED or Severe ED categories as suffering from ED in this analysis. *Among those reporting they had a sexual partner in the last 4 weeks (N = 2,981). Between age band or between center differences P < 0.05 except. Due to rounding percentages may not add up to 100. Non-Trans. = non-transitional centers, Trans. = transitional centers.

10 General and Sexual Health in European Men 1371 Figure 1 Adjusted odds ratio (log scale with 95% confidence intervals) for different sociodemographic parameters associated with erectile dysfunction categorizing yes/no parameters as dummy 0/1. Model additionally adjusted for age, center, and self-reported health. terms of timing of orgasm. There was no significant difference between transitional and nontransitional centers regarding satisfaction with overall sexual relationship, although a higher proportion of men in transitional centers reported that they were dissatisfied or neutral concerning their nonsexual relationship (Table 5). Relationship Between Sociodemographic Parameters and ED Figure 1 shows the association between different sociodemographic parameters and ED. After adjustment for confounding factors, including age, self-reported health, and center, current smoking was a significant risk factor for ED. Employment status played an apparently protective role, while both reported partner s health and satisfaction with sexual relationship were independent risk factors. Relationship Between Morbidities and Sexual Function Figure 2 shows the relationships between CVD, hypertension, diabetes, obesity, LUTS, and depression with sexual function outcomes. Depression was the only factor significantly associated with all sexual function parameters studied. LUTS were associated with ED and orgasm frequency. ED was also influenced by CVD, diabetes, and obesity. Finally, CVD was also a risk factor for the frequency of sexual intercourse. Relationship Between Health/Lifestyle Factors and Health-Related QoL Figure 3 shows the joint contributions of health and lifestyle factors to low QoL (defined as scoring in the lowest quartile of either the SF-36 physical or mental component). In the entire cohort, CVD, depression, LUTS, obesity, ED, and hypertension were independent predictors of low physical QoL. Depression was the only parameter associated with low mental QoL when the entire cohort was evaluated. When transitional (Lodz, Szeged, Tartu) and nontransitional countries (Florence, Santiago, Manchester, Leuven, Malmo) were considered separately, some major differences were observed. CVD, depression, LUTS, and hypertension were associated with low physical QoL in transitional countries. Similar results were observed in nontransitional centers except for the lack of an association with hypertension and a significant association with obesity. Depression was significantly associated with low mental QoL in both nontransitional and transitional countries; LUTS were associated with low mental QoL in transitional countries only. Discussion EMAS is the largest multicenter, population-based study of aging in European men which has allowed us to systematically analyze, for the first time, different aspects of both general and sexual health by

11 1372 Corona et al. Figure 2 Adjusted odds ratio (log scale with 95% confidence intervals) for different sexual parameters associated with morbidities. Adjusted for age, self-rated general health, center, CVD, hypertension, diabetes, obesity, LUTS, and depression categorizing yes/no parameters as dummy 0/1. CVD = cardiovascular diseases; self-reported heart condition and/or history of stroke, hypertension = self-reported high blood pressure and/or using antihypertensive medication; diabetes = selfreported diabetes and/or using antidiabetic medication; obesity = body mass index 30 kg/m 2 ; LUTS = lower-urinary tract symptoms as derived from international prostate symptom score (IPSS bands moderate and severe); depression as derived from Beck s Depression Inventory score mild and/or using antidepressants. standardized methodologies across eight European centers. In addition, we investigated the association of general and sexual health with a subjective measure of QoL. One of the main strengths of the study is that the participating centers were from different regions of Europe, including transitional countries, thus allowing comparisons between regions with different socioeconomic and geopolitical backgrounds. In particular, in transitional countries, the collapse of previous political systems and the lack of economic resources have resulted in a rapid deterioration of population health due to limited access to health care, growing inequity, increasing medication costs, and cutbacks in preventive care [2 7,15]. Accordingly, Jagger et al. [5] demonstrated that both life expectancy and healthy life years are lowest in the transitional countries of the European Union, with excess CVD considered as the most important determinant of the life expectancy gap between eastern and western Europe [2,5,8,15,16]. About half of our subjects were overweight (BMI between 25 and 30) and more than 50% of them reported one or more morbidities associated with age. Overall, hypertension, obesity, and heart diseases were the most common conditions with the highest prevalence observed in transitional countries. Smoking, another risk factor for CVD [8], was also more prevalent in transitional countries (25%) than in nontransitional centers (19%). One-third of subjects described their general health as fair/poor with a higher rate in transitional countries. In particular, Szeged showed the

12 General and Sexual Health in European Men 1373 Figure 3 Association between SF-36 component scores and predictors. Adjusted (age, age left education, smoking and other covariates, categorizing yes/no parameters as dummy 0/1) odds ratio (log scale with 95% confidence intervals) of being in the lowest quartile of SF-36 physical (upper line) and mental (lower line) component summary score, estimated in the entire cohort and separately for European Male Ageing Study region. CVD = cardiovascular diseases: self-reported heart condition and/or history of stroke, hypertension = self-reported high blood pressure and/or using antihypertensive medication; diabetes = self-reported diabetes and/or using antidiabetic medication; obesity = body mass index 30 kg/m 2 ; LUTS = lower-urinary tract symptoms as derived from internal prostate symptom score (IPSS bands moderate and severe); depression as derived from Beck s Depression Inventory score mild and/or using antidepressants. highest prevalence of obesity, hypertension, stroke, and diabetes, while Lodz and Tartu showed the highest prevalence of heart disease and reported the lowest usage of lipid lowering medications [17]. ED is a worldwide condition whose prevalence has been evaluated in different countries. In 1993, the National Institutes of Health Consensus Conference defined ED as the persistent inability to achieve and/or maintain a penile erection adequate for satisfactory sexual activity [18]. Different methods used for the definition of ED, however, have been considered as possible confounding factor for study comparisons [19]. So far, only six studies have simultaneously evaluated the prevalence of ED in different regions of the world using standardized methods [20 25]. The overall mean prevalence of ED ranges from 14% [21] to 48% [20] with a higher prevalence found in the United States and in Southeast Asia when compared with Europe [25]. Overall, 30% of subjects reported ED with a higher prevalence in transitional countries. France and Germany are two big European countries not included in the EMAS survey. Interestingly, Giuliano et al. [26], in a study of 1,004 men (aged <40 years) representative of France s population, reported a similar prevalence of ED 31.6%. Conversely, in Germany, in the Cologne Male Survey, it was found that among 4,489 men (aged years), the prevalence of ED was 19.2% with a steep age-related increase [27]. Organic determinants and, in particular, CVD risk factors have been considered the most important pathogenic factors underlying ED [9 11,25,28 31]. Accordingly, the mean prevalence of ED in transitional countries (36%) was higher than in nontransitional ones (27%). In

13 1374 Corona et al. addition, in our sample, obesity, CVD, diabetes, and smoking represented risk factors for ED even after adjustment for EMAS centers. Interestingly, men in Szeged, reporting the lowest prevalence of ED in transitional countries, also showed a lower prevalence of depression symptoms and a higher employment rate. Hence, other factors (e.g., social, psychological and relational; 28,32), besides organic determinants, may also be important in the pathogenesis of ED, explaining, at least partially, the differences observed among the EMAS centers. Accordingly, we found that the employment rate and perception of partner s health were all independent risk factors for ED. The possibility that different sociocultural backgrounds may determine different reactions to sexual difficulties cannot be excluded. In fact, Perelman et al. [32] reported that attitudes toward ED and behavior relating to the disorder differ among different countries. Accordingly, Chinese and Malaysian men tend to hold their wives responsible for their ED, while Indian men consider ED to be a matter of fate [33]. Lower physical QoL, observed in older subjects, was associated with CVD, depression, LUTS, obesity, ED, and hypertension. Conversely, mental QoL showed no change with aging. Nevertheless, the prevalence of depression was higher in the older age groups. The reasons for this association are probably multifactorial, but poor physical health is considered an important risk factor for depression later in life [34]. EMAS centers in transitional countries, particularly Lodz and Tartu, reported the highest prevalence of depression, the lowest consumption of antidepressant drugs, and the lowest prevalence of healthy subjects. Depressive symptoms in the entire sample, in both transitional and nontransitional countries, were also associated with low mental QoL after adjustment for confounders. Interestingly, the impact of depressive symptoms on mental QoL seems to be higher (HR odds ratio [OR] = 14.9) in nontransitional vs. transitional (HR = 8.48) centers. The reasons for this finding are not clear, but it could be speculated that the severity of various morbidities in transitional countries may render the effect of depression on QoL less relevant. Low QoL in transitional countries was also associated with LUTS, as previously reported [35,36]. The association between LUTS and the mental QoL in transitional countries deserves more clarification. Overall, 27% of respondents reported moderate-to-severe symptoms. Similar data (29%) were previously reported in the Multinational Survey of the Ageing Male, including only western European centers [20]. However, men from transitional countries in EMAS demonstrated the highest prevalence of severe nocturia (>3 times/night), the lowest use of prostate drugs, and the highest prevalence of prostate surgery. Therefore, differences in the management of LUTS between transitional and nontransitional countries may partly explain this finding. ED was the only sexual parameter significantly associated with an impairment of physical QoL, although this relationship was not confirmed when transitional and nontransitional countries were considered separately. ED was not associated with the mental component of SF-36. Data regarding the specific contribution of ED to QoL in the general population are scarce [36]. It could be speculated that the severity of different organic morbidities such as CVD and depression (major determinants of ED as well as both the SF-36 physical and mental component scores) could render any independent effect of ED on QoL less obvious. The frequency of other sexual activities besides erectile function showed some differences. Men in Tartu, who reported poor general health and the highest prevalence of smoking, showed also the lowest prevalence of sexual activities. Conversely, men from Szeged, the most sexually active in the three transitional countries, reported a lower prevalence of heart disease and depression when compared with Lodz and Tartu. However, men from Manchester and Santiago (the least and most sexually active in nontransitional countries, respectively) showed a similar prevalence of concomitant morbidities. Hence, similar to ED, differences in sexual functioning among the EMAS centers cannot be explained by organic determinants, without taking other factors (psychological, relational, or sociocultural) into account. Although the prevalence of ED is higher in older men, concern about it decreased after the sixth decade as previously reported in other crosssectional studies [20,37,38]. Interestingly, similar results were recently reported in the Olmsted County Study of Urinary Symptoms and Health Status among Men involving a random sample of 2,213 men evaluated biennially from 1996 to 2004 [39]. Overall, a decline in all of the sexual function domains (erectile function, libido, ejaculatory function, sexual problems, and sexual satisfaction) was reported; however, significantly smaller correlations between changes in the functional domains

14 General and Sexual Health in European Men 1375 and changes in sexual satisfaction and problem assessment were observed among older men. Sexual activity is lower in older men and in men with ED. Although in our cohort, about 55% complained of ED in the oldest age group ( 70 years), 49% reported at least one sexual intercourse, 24% masturbation, 58% petting, and 75% thinking about sex in the previous 4 weeks, indicating that normal erections are not a prerequisite for the continuation of sexual activity. Accordingly, Perelman et al. [32] reported that more than 40% of men declared that there are other ways to get sexual gratification that do not require a good erection. A similar prevalence of sexual intercourse was recently reported in Swedish men older than 70 years [40]. Satisfaction with sexual function was found to be an independent risk factor for ED. Half of subjects reported that they were satisfied with their sexual relationship, but this fell to 43% in the 70 age group. Conversely, satisfaction with general (nonsexual) relationship was high (more than 80%) and independent of age. In addition, concern about ED and sexual activities was lowest in the oldest age group. Hence, the perception of the importance of ED seems to diminish among the oldest participants, with a significant proportion of men over 70 apparently reconciled to lower levels of sexual activity [37]. Sex remains, however, an important part of elderly people s lives. Overall, men from transitional countries appeared less concerned about ED and more satisfied with their sexual relationships, possibly reflecting attitudinal and/or cultural differences in these societies. Recognizing underlying conditions through ED might be a useful motivation for men to improve their health-related lifestyle choices. The presence of ED, previously considered no more than a frustrating condition, should now be regarded as a unique opportunity to screen for the presence of comorbidities. Hence, ED subjects can be considered in some ways lucky because ED offers them a chance to undergo medical examination and, therefore, to improve not only sexual but also, most importantly, overall health [41]. Depression was the only factor significantly associated with low sexual desire, a relationship that has been well documented [42]. However, our findings suggest that sexual desire is maintained in subjects with ED even in the presence of ED-related morbidities. Similar results have been reported in diabetic patients [43]. Overall, about 6% of subjects reported severe orgasmic impairment, closely associated with aging [20,21,37,38]. Depression and LUTS appeared to be the most important determinants after adjustment for confounders including antidepressive drugs [20,42]. Some limitations should be recognized. The EMAS study did not involve some big European countries such as Germany and France. In addition, the study population in each country was not a truly random sample of the entire population, although it permits some comparison between different areas of European areas. The overall response rate for participation in the study was 41% and it is possible that those who took part may have differed with respect to levels of sexual and general health compared with those who declined to participate. Potential effects of such response bias may be to overestimate or underestimate the true prevalence of sexual dysfunctions and comorbidities within the populations sampled. However, nonresponse is unlikely to influence our results on risk factors and interrelationships between sexual and general health, being based on internal comparisons of responders. Self-reported information in population surveys may be subjected to errors of recall; however, any misclassifications were likely to have been random and the effect, if any, would be to reduce the reported associations toward the null rather than produce spurious associations. Conclusions Our data demonstrate that as men become middleaged and older, they remain sexually active even if sexual dysfunctions associated with comorbidities are more prevalent in older age. In comparison to nontransitional countries, men from the three transitional countries reported the highest prevalence of concomitant morbidities which were associated with a greater impairment of sexual function and lower QoL. Our data also provide new information regarding the complex interrelationships between general and sexual health and overall QoL in aging men. A greater understanding of how these relationships apply to different countries is important in the provision of comprehensive health care for the burgeoning elderly population. In particular, the greater burden and need in transitional countries to develop more effective strategies to promote healthy aging for men is highlighted. Acknowledgments The EMAS is funded by the Commission of the European Communities Fifth Framework Program Quality of Life and Management of Living Resources Grant