Depressive Symptom Severity Is Associated With IPSS and ICIQ‐SF Scores but Not Overactive Bladder in Japanese Male Never‐Smokers

ABSTRACT

Objectives

Lower urinary tract symptoms (LUTS) are common in men and have been linked to mental health, but their relationship with depressive symptoms across the adult age range, including young men, remains unclear. We examined the association between LUTS and depressive symptoms in Japanese men.

Methods

Using an internet research panel, 4000 never‐smoking Japanese men were surveyed. LUTS were assessed with the Overactive Bladder Symptom Score (OABSS), International Consultation on Incontinence Questionnaire‐Short Form (ICIQ‐SF), International Prostate Symptom Score (IPSS), and Quality of Life (QOL) index, and depressive symptoms with the Quick Inventory of Depressive Symptomatology, Japanese version (QIDS‐J). Scores were compared across QIDS‐J severity groups, and multivariable regression was used to adjust for age and household income.

Results

Of 2568 respondents (mean age 46.6 years), more severe depressive symptoms were associated with younger age (p < 0.001). ICIQ‐SF and IPSS increased with QIDS‐J severity (both p < 0.001) and remained independently associated after adjustment for age and income (IPSS adjusted β 0.32; ICIQ‐SF 0.13; both p < 0.001), whereas OABSS was not associated (p = 0.193). The same pattern was observed in men aged 20–40 years.

Conclusions

Greater depressive symptom severity was independently associated with higher IPSS and ICIQ‐SF scores, but not with OABSS, OAB, or UUI among Japanese male never‐smokers.

1. Introduction

Lower urinary tract symptoms (LUTS) comprise conditions such as overactive bladder (OAB), benign prostatic hyperplasia (BPH), and bladder outlet obstruction. In the EpiLUTS survey, LUTS were reported by about 47% of men and 46% of women [1], and other population‐based surveys have estimated a prevalence of roughly 20%–50% in both sexes [2, 3, 4]. The prevalence increases with age, and conditions such as diabetes mellitus, hypertension, and cardiovascular disease have also been linked to LUTS [5, 6, 7, 8, 9].

In men, BPH affects roughly 40% of those aged 50–59 years and 70% of those aged 60–69 years [10], so LUTS are generally more frequent in older men than in women. In men, LUTS have been reported to be influenced by metabolic syndrome, including obesity, diabetes mellitus, dyslipidemia, and hypertension [11, 12]. The EPIC study showed that OAB reduces quality of life (QOL) [13, 14, 15]. Beyond impairing QOL, OAB has also been associated with depression and anxiety [15, 16, 17]. Incontinence and BPH have likewise been reported to correlate with depressive symptoms [18, 19, 20, 21]. Several reports have also indicated that a higher International Prostate Symptom Score (IPSS) is associated with poorer QOL [20, 21, 22]. Few studies have reported the association between LUTS and depression in Asian Japanese men in each generation.

The present study examined the correlation between LUTS and depressive symptoms in each generation. We have previously reported a preliminary analysis of depressive symptoms and overactive bladder in this population [23]; the present study extends that work to multiple validated LUTS instruments across a wide age range.

2. Materials & Methods

Of those registered as monitors at an Internet research company (Freeasy/I‐Bridge), 4000 Japanese males with no smoking history who agreed to “cooperate with the Internet survey, follow the handling of personal information, and use the information as aggregate data” on the response screen of this survey were targeted. Never‐smokers were selected because we have previously identified smoking as a strong factor associated with LUTS in men [24] and with overactive bladder and incontinence in women [25]; restricting the sample to never‐smokers was intended to remove this strong confounder. Four thousand subjects were included in the study. The study period was from the start of the questionnaire to July 30, 2020, two weeks after the start of the survey. The study was approved by the Investigational Review Board (IRB) of Yokohama City University Medical Center (approval number: B201000053). The present study is an expanded analysis of the same cohort as our preliminary report on depression and overactive bladder [23], extended to a substantially larger sample, multiple validated lower urinary tract symptom instruments, and multivariable adjustment. Informed consent to participate was obtained from all respondents before answering the questionnaire. Experimental procedures were conducted in accordance with the ethical standards of the Helsinki Declaration. The raw data used in this study are provided as Supplementary File.

Questionnaires (Overactive Bladder Symptom Score; OABSS, International Consultation on Incontinence Questionnaire‐Short Form; ICIQ‐SF, International Prostate Symptom Score; IPSS, Quality of Life index; QOL index) and depressive symptoms (Quick Inventory of Depressive Symptomatology; QIDS‐J) and information provided by Freeasy Inc. regarding age, household income, living arrangements, occupation, marital status, and presence of children.

Overactive bladder was defined from the four‐item OABSS (daytime frequency, nighttime frequency, urgency, and urge incontinence) as a total of at least 3 together with a Q3 score of at least 2, and urge urinary incontinence (UUI) as a Q4 score of at least 2; daytime frequency corresponded to Q1 of at least 1 and nocturnal frequency to Q2 of at least 2. Incontinence was further characterized with the three‐item ICIQ‐SF (frequency and volume of leakage and its impact on daily life), in which higher totals denote greater severity; within this instrument, UUI was taken from Q4‐2, stress urinary incontinence (SUI) from Q4‐3 or Q4‐5, mixed urinary incontinence (MUI) from the presence of both, and post‐micturition dribble (PMD) from Q4‐6. Voiding symptoms were assessed with the seven‐item IPSS and its QOL item, with nocturia defined as Q7 of at least 1; the voiding subscore (IPSS‐V) was summed from items 1, 3, 5, and 6 and the storage subscore (IPSS‐S) from items 2, 4, and 7, and totals of at least 8 and at least 20 were regarded as moderate and severe, respectively. Depressive symptoms were measured with the 16‐item QIDS‐J, whose severity is derived from a nine‐domain total covering sleep, appetite and weight, psychomotor change, and six further items.

2.1. Statistical Analysis

The participants' characteristics and scores were analyzed by the Mann–Whitney U and one‐factor analysis of variance (ANOVA) tests, and the prevalence of daytime frequency, nocturia, urgency, UUI, SUI, MUI, and PMD was analyzed by a chi‐square test using the Graph Pad Prism software program (Graph Pad Software, La Jolla, CA, USA). p values of < 0.05 were considered to indicate statistical significance.

To assess whether the associations were independent of confounders, multivariable models were additionally constructed with the QIDS‐J total score as the explanatory variable and adjustment for age and household income (categorized as < 4, 4–8, 8–12, and > 12 million JPY). Continuous symptom scores (IPSS, ICIQ‐SF, OABSS, and QOL index) were analyzed by linear regression and the presence of OAB and UUI by logistic regression; regression coefficients (β) or odds ratios (OR) with 95% confidence intervals (CI) were estimated using Python (statsmodels, version 0.14). A two‐sided p value of < 0.05 was considered statistically significant. The analysis code for the multivariable models and the corresponding forest plot (Figure 1) was written with the assistance of Claude (Anthropic), a large language model, which was also used for English‐language editing of the manuscript; all code, outputs, and edits were reviewed and verified by the authors. In an additional, comprehensively adjusted model, we further included marital status, presence of children, and occupation. To examine symptom domains, the IPSS voiding subscore (items 1, 3, 5, and 6), the IPSS storage subscore (items 2, 4, and 7), and nocturia (IPSS item 7) were analyzed separately. Sensitivity analyses were performed excluding the QIDS‐J sleep domain from the exposure, the nocturia item from the IPSS, and the impact item from the ICIQ‐SF.

FIGURE 1.

Multivariable association between QIDS‐J total score and lower urinary tract symptoms, adjusted for age and household income, shown as (a) adjusted regression coefficients (β) for continuous scores (IPSS, ICIQ‐SF, OABSS, and QOL index) and (b) adjusted odds ratios (OR) for binary diagnoses (OAB and UUI), in all men (N = 2568) and in men aged 20–40 years (n = 1056). Estimates are per 1‐point increase in the QIDS‐J total score; horizontal bars indicate 95% confidence intervals.

3. Results

A questionnaire was sent to 4000 Japanese men with no smoking history, and 2568 responded (64.2%). Patient background is shown in Table 1. The mean age was 46.6 ± 15.9 years. Of the 2568 patients, 1232 (48%) were married men and 1008 (39%) had children. Younger men were well represented: 407 respondents (15.8%) were younger than 30 years and 1007 (39.2%) were younger than 40 years, with the remainder distributed across all older age groups (age range 15–95 years).

TABLE 1.

Patients' background.

	total	No‐depression	Mild	Moderate	Severe	Very severe	p
		0–5 points	6–10 points	11–15 points	16–20 points	21‐ points
Number of pts.		1525	535	264	158	86
Age		47 (48.7 ± 16.3)	45 (46.7 ± 15.6)	39 (41.7 ± 13.9)	36 (38.5 ± 12.0)	37.5 (38.3 ± 11.2)	< 0.001
Marriage (yes)		825	244	91	44	28	< 0.0001
Child (yes)		678	202	71	40	17	< 0.0001
Household Income
less than 4000,000 JPY		585	227	133	76	45	0.002
4000,000‐8 000 000 JPY		593	208	95	54	28
8 000 000‐12 000 000 JPY		215	73	21	18	9
more than 12 000 000 JPY		132	27	15	10	4
Working pattern
Civil servant		83	16	10	11	8
Employee		636	241	117	73	39
Free‐lance		35	11	3	0	2
Homemaker		3	2	0	1	0
Manager		35	7	5	5	0
Medical worker		22	4	4	0	0
No occupation		342	102	60	20	15
Others		45	22	7	5	6
Part‐time job		110	53	19	21	5
Self owned work		82	26	9	2	2
Student		38	15	4	9	4
Temporary worker		94	36	26	11	5

Figure 2 shows the mean age in each of the five QIDS‐J categories; age decreased as QIDS‐J severity increased (p < 0.001). Figure 3 shows the mean OABSS, ICIQ‐SF, and IPSS for all ages by QIDS‐J severity. OABSS did not differ significantly (p = 0.254), whereas ICIQ‐SF and IPSS increased significantly with severity (both p < 0.001). Using the standard IPSS bands (0–7 mild, 8–19 moderate, 20–35 severe), the mean IPSS rose from 3.11 in the normal category to 8.59 in the very severe category, moving from the mild into the moderate range, whereas the mean OABSS remained within the mild range (≤ 5) across all categories.

FIGURE 2.

The mean age of each of QIDS‐J: normal, mild, moderate, severe, and very severe.

FIGURE 3.

Mean values of (a) OABS score, (b) ICIQ‐SF score, and (c) IPS score for all age males by QIDS‐J severity.

Figure 4 shows the proportions diagnosed with OAB and UUI for all ages by QIDS‐J severity; the proportion with OAB differed significantly (p = 0.008), whereas UUI did not (p = 0.460). We then focused on men aged 20–40 years (Figure 5). As in the whole sample, OABSS did not differ (p = 0.736), whereas ICIQ‐SF and IPSS increased significantly with QIDS‐J severity (both p < 0.001); the mean IPSS again reached the moderate range in the most severe category, and the mean ICIQ‐SF rose from 0.50 to 4.32.

FIGURE 4.

Percentage of men of all ages diagnosed with (a) OAB and (b) UUI according to QIDS‐J severity.

FIGURE 5.

On men aged 20 to 40, presented the average values of (a) OABS score, (b) ICIQ‐SF score, and (c) IPS score by QIDS‐J severity.

Figure 6 shows the proportions diagnosed with OAB and UUI in men aged 20–40 years; neither differed significantly by QIDS‐J severity (OAB p = 0.708; UUI p = 0.621).

FIGURE 6.

The percentage of men aged 20–40 diagnosed with (a) OAB and (b) UUI according to QIDS‐J severity.

In multivariable analysis adjusting for age and household income, a higher QIDS‐J score remained independently associated with a higher IPSS (adjusted β 0.32 per point, 95% CI 0.27–0.37; p < 0.001) and a higher ICIQ‐SF score (β 0.13, 95% CI 0.11–0.15; p < 0.001), as well as with a poorer QOL index (β 0.05, 95% CI 0.04–0.06; p < 0.001). These associations were slightly stronger after adjustment than in the univariable analysis, indicating that age acted as a negative confounder, since younger men reported more depressive symptoms despite the age‐related increase in LUTS. In contrast, OABSS showed no independent association with the QIDS‐J score (β 0.01, 95% CI −0.01–0.03; p = 0.193). For the binary diagnoses, the crude association between OAB and depressive symptom severity was attenuated to non‐significance after adjustment (adjusted OR 1.02 per point, 95% CI 1.00–1.03; p = 0.082), and UUI was not associated (OR 1.01, 95% CI 1.00–1.03; p = 0.116), suggesting that the univariable OAB difference was largely explained by age. Among men aged 20–40 years, the adjusted associations were consistent, with significant relationships for IPSS (β 0.35, 95% CI 0.28–0.42; p < 0.001) and ICIQ‐SF score (β 0.16, 95% CI 0.12–0.20; p < 0.001) but not for OABSS, OAB, or UUI (Table 2). Adding marital status, presence of children, and occupation to the model (comprehensive adjustment) did not materially change these estimates (IPSS adjusted β 0.33, 95% CI 0.28–0.38; ICIQ‐SF β 0.13, 95% CI 0.11–0.16; QOL index β 0.05, 95% CI 0.04–0.06, all p < 0.001; OABSS β 0.01, p = 0.20; OAB adjusted OR 1.01, p = 0.09; UUI OR 1.01, p = 0.11). When the IPSS was separated into subscores, a higher QIDS‐J score was independently associated with the voiding subscore (β 0.18, 95% CI 0.15–0.21), the storage subscore (β 0.14, 95% CI 0.12–0.16), and nocturia (β 0.04, 95% CI 0.03–0.05) (all p < 0.001); thus, the association was not confined to voiding symptoms (Table S1). In sensitivity analyses, the association of the QIDS‐J score with IPSS and ICIQ‐SF remained significant after excluding the QIDS‐J sleep domain (IPSS β 0.34; ICIQ‐SF β 0.14), after excluding nocturia from the IPSS (β 0.28), and after excluding the impact item from the ICIQ‐SF (β 0.07) (all p < 0.001) (Table S2).

TABLE 2.

Multivariable association between the QIDS‐J depressive symptom score and lower urinary tract symptoms.

Outcome (measure)	Model	Minimally adjusted, all ages (N = 2568)		Comprehensively adjusted, all ages (N = 2568)		Minimally adjusted, age 20–40 (n = 1056)
		Estimate (95% CI)	p	Estimate (95% CI)	p	Estimate (95% CI)	p
IPSS	β	0.32 (0.27 to 0.37)	< 0.001	0.33 (0.28 to 0.38)	< 0.001	0.35 (0.28 to 0.42)	< 0.001
ICIQ‐SF score	β	0.13 (0.11 to 0.15)	< 0.001	0.13 (0.11 to 0.16)	< 0.001	0.16 (0.12 to 0.20)	< 0.001
OABSS	β	0.01 (−0.01 to 0.03)	0.193	0.01 (−0.01 to 0.03)	0.202	0.01 (−0.01 to 0.03)	0.372
QOL index	β	0.05 (0.04 to 0.06)	< 0.001	0.05 (0.04 to 0.06)	< 0.001	0.05 (0.03 to 0.06)	< 0.001
OAB (present vs. absent)	OR	1.02 (1.00 to 1.03)	0.082	1.01 (1.00 to 1.03)	0.086	1.01 (0.99 to 1.03)	0.468
UUI (present vs. absent)	OR	1.01 (1.00 to 1.03)	0.116	1.01 (1.00 to 1.03)	0.111	1.01 (0.99 to 1.04)	0.256

Note: Estimates are the change in outcome per 1‐point increase in the QIDS‐J total score. Minimally adjusted model: age and household income. Comprehensively adjusted model: age, household income, marital status, presence of children, and occupation.

Abbreviations: β, linear regression coefficient (continuous scores); CI, confidence interval; ICIQ‐SF, International Consultation on Incontinence Questionnaire–Short Form; IPSS, International Prostate Symptom Score; OAB, overactive bladder; OABSS, Overactive Bladder Symptom Score; OR, odds ratio (logistic regression; binary diagnoses); QIDS‐J, Quick Inventory of Depressive Symptomatology, Japanese version; QOL, quality of life; UUI, urge urinary incontinence.

4. Discussion

Although there have been long‐term studies evaluating the relationship between lower urinary tract symptoms and depression in healthy men using the IPSS [26], we investigated lower urinary tract symptoms in healthy men using multiple items, including the IPSS, ICIQ‐SF, and OABSS. Although there have been studies that have focused on middle‐aged and older adults [27], our study is the first to also address the association between depressive symptoms and LUTS limited to young adults (20–40 years old), since we obtained results from a wide range of ages, from the 20s on up.

As shown in Figure 3, although no significant difference was observed for OABSS (p = 0.2541), significant differences were observed for ICIQ‐SF score and IPSS (p < 0.0001). Litman et al. reported that depressive symptoms significantly worsen LUTS symptoms and suggested that gender and race are not related factors [7]. However, the details of LUTS symptoms were not described, and the differences between urinary symptoms and storage symptoms remained unclear. Lai et al. focused on OAB (storage dysfunction) and investigated its association with anxiety symptoms, arguing that OAB is strongly associated with severe anxiety symptoms. They also noted that when anxiety symptoms coexist with OAB, storage dysfunction and incontinence symptoms worsen, significantly impairing quality of life. They also argued that this leads to depressive symptoms. Furthermore, there was no gender difference in this regard [17]. In this study, no association was found between OAB and depressive symptoms; and because the adjusted estimates were close to the null, this likely reflects a genuinely small association rather than limited statistical power alone (only 407 adult men, 16%, reported OAB). Regarding Figure 4, the proportion diagnosed with OAB differed significantly across QIDS‐J severity categories (p = 0.008), whereas the proportion with UUI did not (p = 0.460). UUI is one of the symptoms included in OAB. Regarding Figure 5 and Figure 6, the association between QIDS‐J and OABSS, QIDS‐J and ICIQ‐SFscore, and QIDS‐J and IPSS was examined, focusing on men in their 20s to 40s. In Figure 5, the p‐value for OABSS was 0.7358, and only OABSS showed no significant difference. According to Stewart et al., significant OAB symptoms, including UUI, do not appear much until the age of 65 and appear in 8.2% of men aged 65–74 and 10.2% of men aged 75 and older [15]. Therefore, while the number of men reporting OAB symptoms is already low, it decreases further as age decreases. Thus, in Figure 5 and Figure 6, by narrowing the age range of men to the younger group of 20–40 years, the results showed no significant difference in OAB symptoms, including UUI. Additionally, while SEIM et al. reported that the number of men complaining of urinary disorders does not increase significantly until after the age of 60, they also reported that 54.6% of men aged 20–29 met the criteria for mild or more severe IPSS and that this increased to 56.4% among men aged 30–39, indicating that more than half of younger men also reported some form of urinary disorder [9]. For OAB, the adjusted estimate was close to the null, indicating that any independent association with depressive symptoms is likely to be small rather than merely undetected because of the number of OAB‐positive men. On the other hand, men meeting the criteria for mild or more severe IPSS urinary dysfunction were present in large numbers across all age groups (65% of the total), and an association with depressive symptoms was confirmed, resulting in a significant difference. On the other hand, while only 27% of adult men reported some form of incontinence, the results showed that as incontinence symptoms worsened, depressive symptoms also tended to worsen. Although the number of positive cases was not as high as in the IPSS, the fact that a significant difference was observed suggests that LUTS with incontinence is more likely to be accompanied by depressive symptoms than LUTS without incontinence.

As described above, LUTS and depressive symptoms are considered to be strongly related. Our results showed no association between OAB symptoms (urgency and other storage symptoms) and depressive symptoms in men, but we did find an association between urinary dysfunction and incontinence symptoms and depressive symptoms. Depressive symptoms are one of the diseases that are currently on the rise. It is a disease that may lead to suicide as symptoms worsen [28, 29]. While the factors involved are diverse, the association between LUTS and depressive symptoms is not widely recognized. By disseminating such findings, it is desirable to ensure that LUTS treatment, particularly incontinence treatment, is not overlooked and is conducted among general practitioners or through active consultation with urologists.

Notably, the association between depressive symptom severity and the IPSS and ICIQ‐SF scores persisted, and if anything, was strengthened after adjustment for age and household income, whereas the crude association between OAB and depressive symptoms was no longer significant once age was taken into account. Because depressive symptoms were more frequent in younger men, in whom LUTS are generally less prevalent, age operated as a negative confounder. When the IPSS was analyzed by domain, both the voiding and storage subscores, as well as nocturia, were independently associated with depressive symptom severity, indicating that the association reflects overall lower urinary tract symptom burden rather than voiding symptoms specifically; in contrast, the OABSS and the OAB and UUI diagnoses were not independently associated (Table 2). Because the adjusted OAB estimate was close to the null (odds ratio 1.02 per point), any independent association between depressive symptoms and OAB is likely to be small rather than simply undetected owing to limited statistical power.

The direction of this association cannot be determined from a cross‐sectional design, and both pathways are plausible. LUTS and coexisting somatic conditions may impair sleep and quality of life and thereby contribute to depressive symptoms; conversely, depressive and anxiety states may modify bladder afferent signaling and produce sensory urgency—reduced first‐sensation volume and bladder hypersensitivity without detrusor overactivity—a pattern demonstrated urodynamically in psychogenic lower urinary tract dysfunction and reproduced in experimental animals exposed to threat or fear. Both storage and voiding symptoms have been reported in association with depression and anxiety [30]. Urodynamic evaluation may therefore help clarify the temporal sequence in individual patients.

This study has several limitations. First, all information was self‐reported through a web‐based questionnaire, without clinical examination or objective confirmation. Second, whether respondents were currently receiving treatment for LUTS or for depression was unknown, so the findings represent a cross‐sectional snapshot of the current symptom status rather than the untreated disease burden. Third, the cross‐sectional design precludes any inference about the causal direction between LUTS and depressive symptoms. Fourth, although the number of men who met the criteria for OAB was modest, the adjusted estimates were close to the null, so limited statistical power is unlikely to be the sole explanation for the absence of an association. Finally, because smoking is a strong factor associated with LUTS, as we have previously reported [24, 25], the study population was restricted to never‐smokers, which may limit generalizability to the general male population. In addition, several potentially important confounders were not available, including medical comorbidities, obesity, alcohol consumption, sleep disorders, physical activity, medication use, and current treatment for LUTS or depression, so residual confounding cannot be excluded. Because the sample comprised never‐smokers recruited through an internet survey, the findings should be regarded as applying to Japanese male never‐smokers participating in such surveys rather than to the general male population. Finally, some overlap between the depressive‐symptom and urinary‐symptom instruments (e.g., sleep and nocturia items) is possible, although the sensitivity analyses above argue against this being the sole explanation.

Greater depressive symptom severity was independently associated with higher IPSS and ICIQ‐SF scores, but not with OABSS, OAB, or UUI, among Japanese male never‐smokers. Longitudinal studies are needed to clarify the temporal and potentially bidirectional relationship between depressive symptoms and LUTS.

Author Contributions

Masanobu Yamazaki: conceptualization, investigation, writing – original draft. Sahoko Ninomiya: conceptualization, investigation. Akihito Hashizume: supervision. Teppei Takeshima: supervision, data curation. Hiroki Ito: supervision. Kazuhide Makiyama: supervision. Hiroji Uemura: funding acquisition, supervision, writing – review and editing. Jun‐Ichi Teranishi: supervision. Takashi Kawahara: conceptualization, investigation, funding acquisition, writing – original draft, methodology, visualization, formal analysis, software, data curation, supervision.

Funding

The authors have nothing to report.

Ethics Statement

The study protocol was approved by the Institutional Review Board of Yokohama City University Medical Center (approval number B201000053).

Consent

Informed consent was obtained from all participants on the response screen before answering the questionnaire.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1: Association between the QIDS‐J depressive symptom score and IPSS voiding and storage subscores and nocturia.

Table S2: Sensitivity analyses addressing potential item overlap between the depressive‐symptom and urinary‐symptom instruments (all ages, adjusted for age and household income).

Supplementary File (LUTS_depression_rawdata.xlsx). De‐identified participant‐level dataset used to generate the tables and figures.

Data Availability Statement

The raw data to create tables and figures are available as Supplementary File (LUTS_depression_rawdata.xlsx; see Table S1).