Erectile Function and Sexual Activity Are Declining in the Younger Generation: Results from a National Survey in Japan

Abstract

Purpose

Twenty five years have passed since the first national survey on erectile dysfunction (ED) in Japan. The Japanese Society for Sexual Medicine conducted a nationwide survey on the actual status of sexual function targeting men over 20 years old in Japan using validated questionnaires commonly used in clinical practice.

Materials and Methods

Japanese men aged 20 to 79 years participated in our online epidemiological study on sexual dysfunction. Erectile status was assessed by direct questioning and specific questionnaires. Risk factors and frequencies of sexual intercourse, masturbation, nocturnal erections, and feeling sexual desire were assessed. The prevalence of these risk factors was compared between men with and without ED. Prevalence and frequencies were calculated for each 5-year age group. Main outcomes were the prevalence and number of patients with ED and simultaneous evaluation of age-related variations.

Results

Direct questioning of the men revealed that 13.0% felt troubled by ED. Although 81.0% of them had at least some ED symptoms based on a Sexual Health Inventory For Men score of ≤21, the prevalence of men with ED by Erection Hardness Score (EHS), the most appropriate questionnaire for Japanese with low sexual activity, was 30.9%, indicating that 14,012,596 men have ED. Most risk factors were related with ED, whereas frequencies of sexual intercourse, masturbation, nocturnal erections, and feeling sexual desire were affected by aging. However, the low frequency of these factors in the young generation was surprising.

Conclusions

The EHS-based assessment revealed a prevalence of ED of 30.9%, which affected approximately 14 million men, and that the sexual desire, erection stiffness, orgasms, and satisfaction were lower than expected in young Japanese men, especially those aged 20 to 24 years, although those factors tended to worsen with aging. We believe that these findings actually reveal the current sexual status of men in Japan.

INTRODUCTION

Erectile dysfunction (ED) is defined as the persistent inability to attain and maintain an erection sufficient to permit satisfactory sexual performance [1]. As sexual dysfunction, especially ED, significantly reduces quality of life and is linked to various diseases, epidemiological studies on this issue would be highly valuable. According to the well-known Massachusetts Male Aging Study, 52% of men over age 40 years have some degree of ED [2]. Furthermore, another worldwide study reported that three-quarters of men have ED symptoms [3], indicating that ED is not rare and can affect any man. The first official national survey on sexual function in Japan, undertaken in 1998, calculated there to be as many as 11.3 million patients with moderate or complete ED [4]. Furthermore, national surveys on male infertility in Japan conducted in 1996 and 2015 revealed startling data. In the 2015 survey, 13.5% of men experienced male infertility due to ED as the primary cause, a nearly four-fold increase over the 1996 results [5]. This was thought to be due to an increase in the proportion of men in younger reproductive age groups presenting with ED. However, the 1998 national survey on ED did not include men in their 20s, a reproductive age group, nor did it use questionnaires validated internationally.

As 25 years have passed since the last survey and aging is becoming more pronounced in Japan, the prevalence and number of patients with ED are assumed to be increasing. Therefore, the Japanese Society for Sexual Medicine (JSSM) decided to conduct a nationwide survey on the actual status of sexual function targeting men aged ≥20 years in Japan using validated questionnaires such as the International Index of Erectile Function (IIEF), Sexual Health Inventory For Men (SHIM), and Erection Hardness Score (EHS), which are generally used in clinical practice.

MATERIALS AND METHODS

1. Participants

We asked 37,485 men aged 20 to 79 years who were registered with an online research company to participate in an online survey on sexual dysfunction between May 29 and June 24, 2023, and 6,228 (16.6%) agreed to participate. The survey participants were recruited using quota-sampling methods to reflect the actual population structure in Japan. Each quota was predetermined based on the 12 five-year age groups.

2. Questions and questionnaires

We first identified the respondents’ age and asked whether they were married, had children, and what their occupation was. They were then directly asked whether they felt troubled by ED and whether they had any of 13 ED risk factors, including diabetes mellitus, obesity, cardiovascular disease, and others. Their frequencies of sexual intercourse, masturbation, nocturnal erections including morning erection, and feelings of sexual desire also were determined.

The SHIM, EHS, and IIEF questionnaires were used. The SHIM is used to evaluate ED with scoring as follows: 1–7, severe ED; 8–11, moderate ED; 12–16, mild-to-moderate ED; 17–21, mild ED; and ≥22, no signs of ED. Therefore, we rated those with scores below 21 as having at least a few symptoms of ED [6]. The EHS grades erectile hardness on a 5-point scale, with grade 0 indicating no penile growth, grade 1 a large penis but not stiff, grade 2 a stiff penis but not sufficient for insertion, grade 3 a penis stiff enough for insertion but not completely stiff, and grade 4 a completely stiff penis. Grade 3 and above indicate a penis capable of insertion, so sexual activity is not completely impossible; grades 0–2 indicate a penis not capable of insertion, which we judged to be a very distressing ED symptom [7]. Scores for all 15 questions of the IIEF and each subdomain (erectile function, orgasmic function, sexual desire, intercourse satisfaction, and overall satisfaction) were also calculated [8].

3. Methods

First, the prevalence of feeling troubled by ED was assessed by direct questioning, followed by calculating the number and percentage of patients troubled by ED according to the current population distribution. The prevalence of perceived ED and the number of patients troubled by ED in each 5-year age group were also identified. Next, the percentages of the severities of ED were calculated from the SHIM scores, and men with scores ≤21 were considered to have at least some ED symptoms, whether troubled by them or not. The prevalence of ED based on percentage and population distribution of ED was assessed for each 5-year age group. The percentage of patients with EHS grade ≤2 and the prevalence were also assessed at 5-year intervals. Finally, total and sub-domain scores of the IIEF were also evaluated for each 5-year age group.

For risk factors listed in the 3rd edition of the JSSM clinical practice guidelines for ED [9], the respective comorbidity rates for the EHS ≤2 (ED) and EHS ≥3 (no ED) groups were calculated. The frequencies of sexual intercourse and masturbation were rated on a 9-point scale (daily, twice/week, once/week, once every 2 weeks, once/month, once every 2–3 months, once every 6 months, once/year, less than once/year, and not at all), and respective percentages were calculated for each 5-year age group. The frequency of nocturnal erections was rated on a 5-point scale (not at all, less than once/week, about once/week, 2–3 times/week, and daily), and percentages were calculated for each 5-year age group. The frequency of feeling sexual desire was also rated on a 5-point scale (never or little, rarely, sometimes, often, and almost always or always), with percentages calculated for each 5-year age group.

4. Data analysis

The IIEF-15 and sub-scores were analyzed for trends in each 5-year age group. Respective comorbidity rates between the EHS ≤2 and EHS ≥3 groups were analyzed by univariate analysis. p and p-trend values <0.05 were considered statistically significant.

5. Ethics statement

The procedures were approved by the Regional Ethics Committee of Juntendo University (approval number: E22-0452). Written informed consent was obtained from all of the patients.

RESULTS

Participant background data are shown in Table 1. The smallest number of participants (401 men, 6.4%) was between the ages 75–79 years, and the largest group (661 men, 10.6%) was between the ages of 45–49 years. We determined that there were sufficient numbers of participants from each group to allow for adequate analysis. There were 3,795 (60.9%) married men and 3,464 (55.6%) with children. Company employees accounted for the majority of the respondents (51.7%), with 205 students (3.3%) and 1,093 unemployed men (17.5%) also included.

Table 1. Participants' background.

	Age group (y)												Total
	20–24	25–29	30–34	35–39	40–44	45–49	50–54	55–59	60–64	65–69	70–74	75–79
Number (%)	433 (7.0)	442 (7.1)	456 (7.3)	509 (8.2)	558 (9.0)	661 (10.6)	636 (10.2)	526 (8.4)	491 (7.9)	507 (8.1)	608 (9.8)	401 (6.4)	6,228 (100.0)
Married
No	413	316	254	211	222	251	220	174	131	107	89	45	2,433 (39.1)
Yes	20	126	202	298	336	410	416	352	360	400	519	356	3,795 (60.9)
With children
No	420	362	304	258	264	295	272	191	144	118	88	48	2,764 (44.4)
Yes	13	80	152	251	294	366	364	335	347	389	520	353	3,464 (55.6)
Occupation
Civil servant	13	34	30	40	28	51	33	43	24	21	4	0	321 (5.2)
Executive	2	2	2	6	11	21	20	18	28	34	22	11	177 (2.8)
Employee (clerical)	24	64	66	76	94	120	113	76	73	39	17	5	767 (12.3)
Employee (technical)	72	141	133	130	143	162	154	122	93	29	22	8	1,209 (19.4)
Employee (other)	42	109	118	159	173	160	166	130	100	55	25	6	1,243 (20.0)
Self-employed	2	10	12	14	25	46	50	54	52	54	63	40	422 (6.8)
Freelance	2	6	12	9	13	18	16	15	9	13	21	4	138 (2.2)
Stay-at-home husband	0	0	3	1	4	3	3	2	5	5	8	2	36 (0.6)
Part-time job	37	31	24	27	20	23	27	19	31	57	82	25	403 (6.5)
Student	196	5	2	0	2	0	0	0	0	0	0	0	205 (3.3)
Other	12	11	17	13	16	25	15	22	14	24	25	20	214 (3.4)
Unemployed	31	29	37	34	29	32	39	25	62	176	319	280	1,093 (17.5)

Values are presented as number (%) or number only.

The prevalence of men answering “yes” when directly questioned about feeling troubled by ED was only 13.0%, or 5,901,660 men based the population distribution. The prevalence of perceived ED and the number of patients for each 5-year age group are shown in Fig. 1. The rate of men who consider ED bothersome generally tends to increase with age. Surprisingly, SHIM results showed that 52.8% and 12.9% of patients had severe and moderate ED, respectively, with 81.0% of patients having at least some ED symptoms as indicated by a SHIM of ≤21 (Fig. 2A). The calculated population distribution yielded an astonishing 36,582,925 men affected by ED (Fig. 2B). Finally, the prevalence of patients with EHS grade ≤2 for sexual dysfunction was 30.9%, indicating that 14,012,596 men had ED (Fig. 3). Although the prevalence of ED increased with age, it was higher in the 20s than in the 30s and 40s age groups (Fig. 3). Impressively, although IIEF scores decrease with age (p-trend <0.01; Fig. 4), total IIEF and all sub-domain scores were clearly lower in the 20–24-year age group than in the 25–29-year age group. When the 13 risk factors for ED were compared between the two EHS groups, comorbidities were significantly higher in the EHS ≤2 group for all factors other than mental illness (Table 2).

Fig. 1. The prevalence and number of patients with erectile dysfunction as evaluated by the prevalence of erectile dysfunction (ED) estimated from the question “Do you feel troubled by ED”.

Fig. 2. Graphs of erectile dysfunction. (A) Distribution of erectile function as evaluated by the Sexual Health Inventory for Men. (B) The prevalence and number of patients with erectile dysfunction as evaluated by the Sexual Health Inventory For Men.

Fig. 3. Graph showing the prevalence and number of patients with erectile dysfunction as evaluated by the Erection Hardness Score.

Fig. 4. The relation between the total and sub-scores of the International Index of Erectile Function and aging. (A) Total score, (B) erectile function sub-domain, (C) orgasmic function sub-domain, (D) sexual desire sub-domain, (E) intercourse satisfaction sub-domain, and (F) overall satisfaction sub-domain.

Table 2. Risk factors.

	EHS ≤2	EHS ≥3	p-value	Odds ratio	95% confidence interval
					Lower	Upper
Diabetes	17.8	6.7	<0.001	3.04	2.513	3.677
Obesity	26.4	21.0	<0.001	1.35	1.161	1.569
Cardiovascular disease	7.7	2.9	<0.001	2.748	2.085	3.623
Hypertension	38.4	21.0	<0.001	2.352	2.045	2.704
Hyperlipidemia	12.7	10.1	0.012	1.292	1.057	1.58
Smoking (including history)	47.0	42.0	0.002	1.228	1.077	1.401
Chronic kidney disease	3.2	1.2	<0.001	2.741	1.802	4.171
Lower urinary tract symptom	16.5	6.0	<0.001	3.093	2.538	3.769
Neurological disease (cerebral infarction or hemorrhage, Parkinson disease, etc.)	5.1	2.0	<0.001	2.648	1.895	3.701
Pelvic trauma, Spinal cord injury	2.4	1.4	0.011	1.785	1133	2.814
Surgery in the pelvis	4.6	1.0	<0.001	4.656	3.145	6.891
Mental illness (depression, anxiety disorder, panic disorder, etc.)	7.0	7.0	0.974	1.004	0.777	1.298
Sleep apnea syndrome	12.4	8.7	<0.001	1.49	1.213	1.829

EHS: Erection Hardness Score.

Based on the Japanese Society of Sexology definition of sexlessness as “the absence of consensual sexual intercourse or sexual contact by the couple for one month or more in the absence of special circumstances, and the expectation that this condition will continue for a long period of time,” 70.4% of those surveyed have been found to be sexless [10]. Most striking in the analysis of the frequency of sexual intercourse was the marked decrease with age. The percentage of respondents answering “zero” for this question was highest in the 20–24-year age group, exceeding 40%, and the percentage decreased with increasing age (Fig. 5A). However, for the frequency of masturbation, the percentages for every day and twice/week were highest in the 20–24-year age group (Fig. 5B). The percentage of men without nocturnal erection increased with age, whereas that of men with nocturnal erection 2–3 times/week or every day tended to decrease with age. However, the percentage of men with no erections was higher in those in their 20s versus their 30s and 40s (Fig. 6A). Similarly, although the percentage of men who never or rarely felt sexual desire increased with age, the percentage was higher in the 20–24-year age group than those in the next three older age groups (Fig. 6B).

Fig. 5. The relation between the frequency of sexual intercourse (A) and masturbation (B) and aging.

Fig. 6. The relation between the frequency of nocturnal erection (A) and feeling sexual desire (B) and aging.

DISCUSSION

We conducted a nationwide survey on male sexual dysfunction in Japan for the second time in 25 years under the auspices of the JSSM. The age-adjusted prevalence (40–70 years) of moderate ED (may achieve and maintain an erection sufficient for satisfactory intercourse) or complete ED (cannot achieve and maintain an erection sufficient for satisfactory intercourse) was reported to be 15.5% in Brazil, 17.2% in Italy, and 22.4% in Malaysia, but was highest in Japan at 34.5% [11]. A previous retrospective epidemiologic study of eight countries reported a prevalence of ED of 45.2% among men aged 40 to 70 years by self-reported survey, with Italy having the highest (52.2%) and Brazil the lowest (42.1%) [12]. An Internet survey similar to ours of around 12,490 men in the United Kingdom showed a prevalence of 41.5% for self-reported ED [13] whereas our study showed a very low self-reported prevalence of ED of 13.0%. However, when directly questioned about whether our participants considered ED as bothersome, they may not have answered “yes” because they did not care about ED, even if they had it, as they had no opportunity for sexual intercourse.

The prevalence of ED varies considerably from survey to survey even by the evaluation of the SHIM: the Netherlands, 16.8% [14]; USA, 20.7% [15]; Australia, 40.3% [16]; New Zealand, 42% [17]; Nigeria, 58.9% [18]; and Poland, 61.1% [19]. Prevalence in Asia is lowest in Taiwan (26.0%) [20] but is relatively high in China (40.56%) [21], Singapore (51.3%) [22], and Malaysia (71.2%) [23]. Our SHIM analysis showed a surprisingly high prevalence of 81.0%. However, if a man has not had sexual intercourse in the last 6 months, he is likely to have a low score on the SHIM because he would have scored 0 for 4 of the 5 questions. In this sense, evaluation by EHS, which assesses erections during sexual intercourse, nocturnal erectile phenomena, and erections during masturbation, seemed most appropriate for the less sexually active Japanese. The EHS clearly showed a prevalence rate of 30.9% for men with ED who are not capable of insertion, i.e., about 14.0 million men. In light of the 1998 survey’s assumption of about 11.3 million patients with ED and the subsequent aging of the population, our figure of about 14.0 million ED patients may be realistically acceptable. Furthermore, the data spurring the relationship between the prevalence of ED by defining it as an EHS score ≤2 and age showed the prevalence to be higher in those in their 20s versus those in their 30s and 40s. In fact, the prevalence rate in the 20–24-year age group was 26.6%, almost equal to that among the 50–54-year age group (27.8%). Furthermore, this serious problem in young men was also apparent in the evaluation by IIEF scores (Fig. 4), although IIEF is intended for sexually active men. The spreading of ED among this generation may be related to the recent enrichment of the Internet environment and social media. Several Internet surveys have shown that men who have watched sexual videos since their teens are more likely to have ED and that ED is more common among those with a strong habit of watching sexual videos [24]. Members of so-called generation Z, which have grown up with the development of the Internet and smartphones, are called “digital natives” or “social networking service natives.” They can easily watch sexual videos, and many young people may have watched them for a long time. This could clearly contribute to the high prevalence of ED among Japanese youth.

When evaluating risk factors for ED, the prevalence of metabolic factors including diabetes, obesity, hypertension, and hyperlipidemia; other systemic diseases including cardiovascular, chronic kidney, and neurological disease, and lower urinary tract symptoms; pelvic trauma and surgery; and smoking and sleep apnea syndrome was significantly higher in the EHS ≤2 group than EHS ≥3 group. The association of these factors was fully predictable even before the survey. Contrary to expectations, however, there was no difference in the prevalence of mental illnesses such as depression and anxiety between the two groups. Generally, mental illness is closely associated with erectile function because erectile strength can be easily impaired by mental disturbance from social stress, pressure at home, and anxiety about work and future. Indeed, although a previous cross-sectional study showed depression to be a significant risk factor [11], that type of survey encourages subjects to respond directly, and the methodology is different from our Internet survey. Recently, another Internet survey found depression and anxiety to be risk factors for ED [13] by showing respective rates of comorbidity for depression and anxiety of 24.3% and 23.3% in ED patients. In the present national survey, however, the rate of mental illness was only 7.0%, which is extremely low compared to previous reports. As our survey encouraged free participation, perhaps those with these mental conditions were not even active enough to participate in the Internet survey. Likely, only those with very mild symptoms participated, and that few people with mental illness actually participated in the survey may be one of the reasons for the similar comorbidity rates in the two groups.

Surprisingly, the percentage of study participants reporting sexlessness (intercourse less than once/month) was almost 70.4%. This percentage tended to increase with age and exceeded 80% for those >70 years old. Interestingly, however, this percentage exceeded 60% in the 20–24-year age group, higher than the percentages for the three groups comprising 25–39-year-olds. Similar to the higher percentages of sexlessness, the results suggest that the 20–24-year age group may be less sexually active and may be additional evidence that sexual activity among Japanese youth is declining. However, masturbation was clearly more frequent among the younger age groups. This indicates that Japanese youth, while not uninterested in sex, may be less active and reluctant to engage in sexual activity with partners and may simply prefer virtual sexual stimulation via the Internet and social networking services. More than 50% of the participants in the 20–24-year age group were aware of nocturnal erections occurring at least once every two days, but their prevalence declines with age, reaching as low as 15% for those over age 70. As noted in the various guidelines for late-onset hypogonadism, lower serum testosterone levels in aging men may be one factor that reduces the frequency of nocturnal erections [25]. This can also decrease libido as was evident in our survey. Libido in the 20–24-year age group was lower than that in the two groups spanning ages 25 to 34 years and almost the same as that in the 35–39-year age group.

This study has several limitations. First, because the survey was conducted via the Internet, credibility of the responses may be of concern. However, with the current proliferation of information gathering via the Internet, it is difficult to collect data on a large number of cases without using this method. As similar surveys are conducted via the Internet in other countries, we consider this to be acceptable. Second, the valid response rate was low, partly due to the short survey period. In a normal Internet-based survey, the number of survey targets is often secured by increasing the number of people requested to participate in the survey. However, we tried to increase the response rate as much as possible by requesting the same person multiple times. Third, even if young people have sexual partners, their sexual activity may not align with their marital status. We attempted this type of analysis out of concern that an analysis using only married respondents would result in a small sample size. Finally, because the prevalence of ED determined by direct questioning was so low, we considered that the participants might not have properly understood the definition and status of ED. However, we believe that these factors are also characteristic of the Japanese perception of sexuality.

CONCLUSIONS

The present study identified the current prevalence of ED in Japan. The self-reported rate of ED was shockingly low at 13.0%, yet SHIM-based assessment indicated that 81.0% of the men had at least some ED symptoms. However, the EHS-based assessment, considered the most appropriate for the sexual life situation in Japan, revealed a prevalence of 30.9%, which translates to approximately 14 million people. Further, the rate of sexlessness was as high as 70.4%. This study also revealed that although young Japanese men, especially those aged 20 to 24 years, frequently masturbate, their sexual desire for others is declining, and not only the stiffness of their erections, but also their orgasms and satisfaction are lower than expected.

Data Sharing Statement

The data required to reproduce these findings cannot be shared at this time as the data also forms part of an ongoing study.