Symptom Bother Is Associated With Medical Care‐Seeking for LUTS in Japanese Women: A Survey of 1800 Women

ABSTRACT

Background and Objective

Lower urinary tract symptoms (LUTS) are highly prevalent in women, yet care‐seeking is very low, and a large treatment gap remains despite the availability of effective treatments. Previous studies, mainly from Western populations, suggest that symptom bother, as well as symptom severity, is associated with medical care‐seeking; however, quantitative data in Asian and particularly Japanese women are scarce, the influence of bother on the separate care indicators is unclear, and the willingness to pay (WTP) for treatment has rarely been quantified. We therefore examined the factors associated with medical care‐seeking, including symptom bother, and the WTP, in Japanese women.

Methods

We conducted a nationwide web‐based survey of 1800 Japanese women aged 40 years or older, covering the OABSS, ICIQ‐UI SF, symptom bother (0–10), separate care indicators (consulting, visiting, treatment, and current treatment), reasons for not seeking care, and WTP. Factors associated with visiting a clinic were assessed with multivariable logistic regression.

Results

Among symptomatic women, only 12.9% had visited a clinic. Both symptom bother and measured symptom severity were independently associated with visiting (bother ≥ 5, adjusted OR 5.19, 95% CI 3.08–8.76; OABSS ≥ 6, adjusted OR 2.24, 95% CI 1.23–4.07), while age, marital status, children, housing, and income were not significant. The most common reason for not seeking care was “thought it was just aging” (65.1%), and WTP rose with bother.

Conclusions

In Japanese women, symptom bother was independently associated with medical care‐seeking (visiting a clinic) for LUTS, as was symptom‐score severity.

1. Introduction

Lower urinary tract symptoms (LUTS) are common and greatly lower quality of life (QOL) through symptoms such as frequency, urgency, nocturia, and urinary incontinence. In the EPIC study of five countries, the prevalence of overactive bladder (OAB) in adults was 11.8% [1], and the nationwide JaCS 2023 survey reported that the prevalence of LUTS in people aged 40 years or older in Japan was 82.5% (women 80.1%) [2]. However, in contrast to this high prevalence, the care‐seeking rate is very low. In JaCS 2023, only 9.9% of women with OAB symptoms were currently receiving treatment [2]. Although effective treatments are established [3, 4], many women do not receive them, and a large treatment gap exists.

A key concept to understand this treatment gap is symptom “bother.” Traditionally, care‐seeking was thought to be explained by symptom severity. However, the EPIC study and later studies showed that how much a patient feels bothered by the symptoms is associated with care‐seeking independently of the measured severity of storage symptoms and urinary incontinence (OABSS and ICIQ‐UI SF, which mainly capture storage and incontinence symptoms rather than voiding symptoms) [5]. In the US EpiLUTS study, bother, the number of symptoms, and age were related to the motivation to seek care, and bother in particular was an important predictor of treatment‐seeking [6]. In other words, both the “amount” of symptoms and the degree of subjective “bother” caused by them may be associated with whether a woman visits a clinic.

However, this knowledge about bother still has several important limitations. First, most of the evidence, including EPIC and EpiLUTS, comes from Western populations, and quantitative data on care‐seeking in Asian people, especially Japanese women, are scarce [7]. The prevalence, symptom pattern, and even the sex difference in bother of OAB may differ between Asia and Western countries [7], so Western findings may not directly apply to Japanese women. Second, the relationship between bother and care‐seeking has mostly been evaluated at a single point, that is, whether a person visited or not, and it is not clear how bother relates to the separate care indicators—consulting, visiting, treatment, and current treatment. Third, the background of not seeking care (such as normalization by aging, stigma, or worry about cost) and the willingness to pay (WTP), which can quantify the potential demand for treatment, have rarely been measured in previous bother studies, so the relationship between bother and economic evaluation is unknown. Therefore, clarifying bother and these other background factors, as well as symptom severity, is important to decide which women should be targeted for awareness and encouragement to seek care.

To address these issues, we conducted and analyzed a large web‐based survey of symptom bother and barriers to care‐seeking for urinary symptoms in Japanese women.

2. Materials and Methods

2.1. Study Design and Participants

This was a cross‐sectional study that used a nationwide web‐based survey. The participants were Japanese women aged 40 years or older. We recruited them from the registered members of an internet survey panel (Freeasy; iiBRIDGE, Tokyo, Japan). The survey was distributed to members of a Japanese internet research panel of approximately 13 million registered monitors (Freeasy; iiBRIDGE). Responses were collected until the predefined target of 1800 completed responses per age stratum quota was reached, after which the survey closed automatically; the number of invitations sent is therefore not available from the platform. A total of 1800 women completed the survey, and because all items were mandatory there were no incomplete responses. Respondents came from all 47 prefectures of Japan. Because a population‐proportional design does not recruit a sufficient number of older women in web‐based surveys, as we observed in our previous studies [8, 9], we recruited participants by age stratum: women in their 40s and 50s were set at 600 each, and women in their 60s (n = 396) and 70 years or older (n = 204) were also recruited. Accordingly, the age distribution differs from that of the general population by design. Their geographic distribution closely matched the 2024 female population of Japan (Table S1). All questionnaire items were mandatory; therefore, there were no missing responses, no participants were excluded, and all 1800 women were analyzed. This study was conducted in accordance with the Declaration of Helsinki. The study was approved by the Institutional Review Board of Yokohama City University Medical Center, Yokohama, Japan (F260700016). On the first page of the questionnaire, participants were shown an explanation of the study and asked to confirm their consent. Only those who gave electronic informed consent then proceeded to answer the questionnaire.

2.2. Questionnaire and Definitions

The questionnaire asked about basic characteristics (age, prefecture, marital status, having children, housing, and household income), urinary symptoms, care‐seeking behavior, reasons for not seeking care, current self‐management, and WTP. We measured symptom severity with two validated tools. The Overactive Bladder Symptom Score (OABSS) has four items and ranges from 0 to 15. The International Consultation on Incontinence Questionnaire‐Urinary Incontinence Short Form (ICIQ‐UI SF) was used for urinary incontinence, and we calculated the sum of the frequency, amount, and quality‐of‐life items. OAB was defined as an OABSS of 3 or more with an urgency item score of 2 or more. Urinary incontinence was defined as an ICIQ‐UI SF frequency item of more than 0. Symptom bother was measured with a single item that asked how much the woman was bothered by her urinary symptoms, on a scale from 0 to 10. For the analysis, we divided bother into three groups: None (0), Low [1, 2, 3, 4], and High [5, 6, 7, 8, 9, 10]. In the multivariable model, high bother was defined as a score of 5 or more. The bother item used an 11‐point numerical rating scale (NRS): “Overall, how bothersome do you find your current urinary symptoms?” (0 = not bothered at all, 10 = unbearably bothered), with the same wording as our companion study of men to enable future comparison between the sexes. A single‐item NRS is a widely accepted method for subjective experiences such as pain; its correlation with the validated OABSS and ICIQ‐UI SF scores is reported in Section 3. This a priori grouping was pragmatic and intended to show trends rather than to serve as a validated classification; the cut‐offs reflect the distribution of the score (Figure S1), in which 0 was the modal value and 5 is the midpoint of the 0–10 scale. Because the score was strongly right‐skewed, bother was also analyzed as a continuous variable, with consistent results. We defined five care indicators, analyzed as separate cross‐sectional responses rather than a sequential pathway: symptomatic (aware of urinary symptoms), consulted (talked to anyone about the symptoms), visited (visited a medical institution), treated (received treatment), and ongoing treatment (currently receiving treatment). WTP was assessed with the question, “If your current urinary symptoms were to improve, how much would you be willing to pay per day?”, and responses were recorded in Japanese yen (JPY) per day. To reduce the influence of extreme values, WTP was winsorized at JPY 5000 (approximately the 99th percentile) for the calculation of means.

2.3. Statistical Analysis

We described categorical variables as numbers and percentages, and continuous variables as means or medians. For the visit rate at each symptom‐severity category, we calculated 95% confidence intervals with the Wilson method. For the figures that show visit rate by raw score, we added a logistic‐regression fit line. To find the factors that were independently related to visiting a clinic, we used a multivariable logistic regression. The outcome was whether the woman had visited a medical institution. All explanatory variables were binary: OABSS of 6 or more, age of 65 years or more, bother score of 5 or more, married, having children, home ownership, and household income of 5 million yen or more. The model included 93 events, which gave about 13 events per variable. We reported adjusted odds ratios (OR) with 95% confidence intervals. For WTP, we calculated the proportion of women willing to pay and the mean WTP after winsorizing at 5000 yen to reduce the effect of extreme values. A two‐sided p‐value of less than 0.05 was considered significant. All analyses were performed with Python (pandas, NumPy, SciPy, and statsmodels). Figures were drawn with Matplotlib.

2.4. Use of Generative AI

The authors used a generative AI tool (Claude, Anthropic) to assist with writing code for the figures and with English‐language editing and proofreading. The AI tool was not used to collect data, to decide the analysis, or to interpret the results. The authors checked all outputs and take full responsibility for the content of this article.

3. Results

3.1. Participant Characteristics

The characteristics of the 1800 participants are shown in Table 1. Age, marital status, having children, housing, and household income did not differ between women who visited a medical institution and those who did not. In contrast, the OABSS total score, the ICIQ‐UI SF total score, and the bother score were all significantly higher in women who visited a medical institution (all p < 0.001) (Table 1). The symptom bother score correlated significantly with the validated OABSS (Spearman's ρ = 0.53) and ICIQ‐UI SF (ρ = 0.69) scores (both p < 0.001). The geographic distribution of respondents across the 47 prefectures closely matched the 2024 female population of Japan (Spearman's ρ = 0.89; Pearson's r = 0.98 for the percentage share; both p < 0.001; Table S1).

TABLE 1.

Characteristics of participants according to whether they visited a medical institution.

Characteristic	Total (n = 1800)	Visited (n = 93)	Not visited (n = 1707)	p
Age, years	54 (47–63)	52 (47–63)	54 (47–62)	0.839
40–49 years	600 (33.3)	31 (33.3)	569 (33.3)	1.000
50–59 years	600 (33.3)	29 (31.2)	571 (33.5)	0.735
60–69 years	396 (22.0)	20 (21.5)	376 (22.0)	1.000
≥ 70 years	204 (11.3)	13 (14.0)	191 (11.2)	0.401
Married	1129 (62.7)	54 (58.1)	1075 (63.0)	0.378
Has children	994 (55.2)	51 (54.8)	943 (55.2)	1.000
Homeowner	1253 (69.6)	58 (62.4)	1195 (70.0)	0.132
Household income ≥ JPY 5 million/year	866 (48.1)	40 (43.0)	826 (48.4)	0.338
OABSS total score	1 (0–3)	3 (2–6)	1 (0–3)	< 0.001
OABSS ≥ 6	132 (7.3)	25 (26.9)	107 (6.3)	< 0.001
ICIQ‐UI SF total score	0 (0–3)	2 (0–9)	0 (0–2)	< 0.001
Overactive bladder	208 (11.6)	34 (36.6)	174 (10.2)	< 0.001
Any urinary incontinence	402 (22.3)	45 (48.4)	357 (20.9)	< 0.001
Bother score (0–10)	0 (0–1)	3 (0–6)	0 (0–1)	< 0.001
Bother score ≥ 5	185 (10.3)	38 (40.9)	147 (8.6)	< 0.001
Aware of urinary symptoms	402 (22.3)	52 (55.9)	350 (20.5)	< 0.001
Willingness to pay, JPY/day	10 (0–100)	100 (10–1000)	10 (0–100)	< 0.001

Note: Data are median (interquartile range) for continuous variables and n (%) for categorical variables. p‐values were calculated with the Mann–Whitney U test for continuous variables and Fisher's exact test for categorical variables. ICIQ‐UI SF, International Consultation on Incontinence Questionnaire‐Urinary Incontinence Short Form; JPY, Japanese yen; OABSS, Overactive Bladder Symptom Score.

3.2. Symptom Score and Clinic‐Visit Rate

We examined the relationship between the urinary symptom scores (OABSS and ICIQ‐UI SF) and the clinic‐visit rate. In the low‐score groups, the visit rate was low and the confidence intervals were narrow (OABSS 0–2, 2.7%; 3–5, 9.0%; ICIQ‐UI SF 0, 3.4%; 1–5, 4.1%). As the score became higher, the visit rate increased, but the confidence intervals also became wider. This means that women with a high symptom score did not always visit a clinic, and that they were divided into those who visited and those who did not (Figure 1, Figure S2).

FIGURE 1.

Clinic‐visit rate by symptom‐severity category. Percentage of women who visited a clinic, by OABSS category (A) and ICIQ‐UI SF category (B). Dots show the observed rate, and error bars show 95% confidence intervals. The number of women (n) is shown under each category. The visit rate increased as the symptom score became higher.

3.3. Care Indicators and Symptom Bother

Among women who were aware of urinary symptoms (n = 402), and reporting these as separate cross‐sectional indicators rather than a strictly sequential pathway, 13.2% had consulted anyone about the symptoms, 12.9% had visited a medical institution, 10.2% had received treatment, and 6.0% were currently receiving treatment. Some women had visited a clinic without first consulting anyone (n = 20), so consulting and visiting did not form a strict sequence, and treatment and current treatment were by definition subsets of clinic visiting. In this process, women with high bother had a higher proportion at each indicator than women with no or low bother, and the visit rate differed significantly among the three groups (None 11.5%, Low 5.9%, High 24.1%; chi‐squared test, p < 0.001). In particular, the treatment rate was significantly higher in the high‐bother group than in the no‐bother group (20.4% vs. 6.6%, p = 0.013) (Figure 2, Table 2).

FIGURE 2.

Care indicators by symptom bother. Percentage of symptomatic women meeting each care indicator (symptomatic, consulted, visited, treated, and currently treated), shown as separate cross‐sectional proportions rather than a sequential pathway. Women are divided into three bother groups by the bother score (Q16, 0–10): None (0), Low (1–4), and High (5–10). Women with high bother met each care indicator more often than the other groups.

TABLE 2.

Characteristics and care‐seeking according to symptom bother group.

Characteristic	None (n = 1139)	Low (n = 476)	High (n = 185)	p
Age, years	53 (47–61)	56 (49–64)	54 (47–64)	0.002
Married	713 (62.6)	301 (63.2)	115 (62.2)	0.958
Has children	605 (53.1)	278 (58.4)	111 (60.0)	0.058
Homeowner	795 (69.8)	341 (71.6)	117 (63.2)	0.106
Household income ≥ JPY 5 million/year	552 (48.5)	230 (48.3)	84 (45.4)	0.738
OABSS total score	1 (0–2)	2 (1–4)	5 (2–8)	< 0.001
OABSS ≥ 6	7 (0.6)	51 (10.7)	74 (40.0)	< 0.001
ICIQ‐UI SF total score	0 (0–0)	3 (0–5)	8 (0–11)	< 0.001
Overactive bladder	24 (2.1)	93 (19.5)	91 (49.2)	< 0.001
Any urinary incontinence	46 (4.0)	233 (48.9)	123 (66.5)	< 0.001
Aware of urinary symptoms	61 (5.4)	204 (42.9)	137 (74.1)	< 0.001
Consulted anyone	37 (3.2)	27 (5.7)	36 (19.5)	< 0.001
Visited a medical institution	29 (2.5)	26 (5.5)	38 (20.5)	< 0.001
Received treatment	7 (0.6)	16 (3.4)	30 (16.2)	< 0.001
Currently receiving treatment	3 (0.3)	10 (2.1)	14 (7.6)	< 0.001
Willingness to pay, JPY/day	0 (0–100)	100 (5–112)	100 (50–500)	< 0.001

Note: Bother groups were defined by the bother score (0–10): None = 0, Low = 1–4, High = 5–10. Data are median (interquartile range) for continuous variables and n (%) for categorical variables. p‐values were calculated with the Kruskal–Wallis test for continuous variables and the chi‐squared test for categorical variables. Abbreviations are as in Table 1.

3.4. Factors Associated With Care‐Seeking

In the multivariable logistic regression for visiting a clinic, age, marital status, having children, housing (owned or rented), and household income were not significant factors. In contrast, OABSS of 6 or more (adjusted OR 2.24, 95% CI 1.23–4.07) and a bother score of 5 or more (adjusted OR 5.19, 95% CI 3.08–8.76) were each independent factors associated with visiting (Figure 3).

FIGURE 3.

Adjusted predictors of care‐seeking. Adjusted odds ratios (OR) and 95% confidence intervals for visiting a clinic, from a multivariable logistic regression. All variables are binary. An asterisk (*) means p < 0.05. Higher OABSS (≥ 6) and high bother (≥ 5) were independently related to visiting. Age and marital status were not significant.

3.5. Reasons for Not Seeking Care

Among women who did not visit a clinic (n = 350), the most common reason was “thought it was just aging” (65.1%), followed by “not bothersome enough to visit” (38.9%), “self‐managed” (18.0%), “worried about cost” (13.4%), and “embarrassed” (13.1%) (Figure 4).

FIGURE 4.

Reasons for not seeking care. Percentage of symptomatic women who did not visit a clinic and selected each reason (multiple answers allowed). The most common reason was “thought it was just aging.”

3.6. Willingness to Pay and Bother

The WTP for symptom relief increased with bother. The proportion of women willing to pay was 38% in the no‐bother group, 77% in the low‐bother group, and 87% in the high‐bother group, and the mean WTP (winsorized at JPY 5000) also increased (JPY 160, 262, and 631, respectively; approximately USD 1.1, 1.7, and 4.2 at the 2024 mean rate of JPY 151.45 per USD) (Figure 5).

FIGURE 5.

Willingness to pay for symptom relief, by bother. (A) Percentage of women who were willing to pay more than 0 yen. (B) Mean willingness to pay (WTP) per day, winsorized at 5000 yen, with standard error. Both the proportion and the amount increased as the bother became higher.

4. Discussion

In this study, we used a nationwide web‐based survey of 1800 Japanese women aged 40 years or older and examined care‐seeking for LUTS, its determinants, and the WTP. Only 12.9% of women who were aware of urinary symptoms had ever visited a medical institution. The visit rate increased step by step as the OABSS severity category became higher, from 2.7% (mild) to 26.8% (severe), but it stayed limited in every category. JaCS 2023, a nationwide survey, also reported that only 9.9% of women with OAB symptoms are currently receiving treatment [2]. This low care‐seeking is not only seen in Japan but is reported consistently in other countries. In the EPIC study and the US EpiLUTS study, less than 38% of women with urinary incontinence contacted their general practitioner [5], and a study in the Middle East reported that about 80% of women had never sought care. Beyond the differences in definitions and health‐care systems, this independent cohort again confirmed that care‐seeking and treatment for female LUTS are low worldwide. A previous study that used the JaCS 2023 data evaluated WTP and symptom clusters, but it did not evaluate the factors associated with care‐seeking [10], and our study adds this point.

The main finding of this study is that medical care‐seeking (visiting a clinic) was independently associated with the patient's own symptom bother, as well as with symptom‐score severity; because these were measured on different scales and dichotomized at different thresholds, the OR should not be compared directly to rank their relative importance. In the low‐bother group, the proportions at each indicator were low (consulted 6.8%, visited 6.2%, treated 3.6%, currently treated 2.6%), while in the high‐bother group it stayed higher at every step (consulted 22.6%, visited 24.1%, treated 20.4%, ongoing 10.2%). So higher bother was associated not only with visiting a clinic but also with a higher proportion having received treatment; reasons for stopping treatment were not collected, so the ongoing‐treatment proportion is not interpreted as treatment continuation. In the multivariable analysis, OABSS of 6 or more (adjusted OR 2.24, 95% CI 1.23–4.07) and a bother score of 5 or more (adjusted OR 5.19, 95% CI 3.08–8.76) were each independently associated with visiting; because these were measured on different scales and dichotomized at different thresholds, the OR should not be compared directly to rank their relative importance. On the other hand, age, having children, housing, and household income were not significantly related. The finding that bother is independently associated with care‐seeking has also been reported in the EPIC study on OAB [5] and in a path analysis of patients with Type 2 diabetes and OAB [11]. In particular, Fitzgerald et al. followed parous women for a long time and showed that the bother score forms a clear threshold that predicts care‐seeking [12], which agrees with our finding of a step‐by‐step increase in the visit rate across OABSS categories. Our study newly showed that these relationships also hold in Japanese women, and that they were consistent across the separate care indicators examined. An important point is that effective treatments for OAB and urinary incontinence are already established in the European (EAU) [3] and US (AUA/SUFU) [4] guidelines, which include behavioral therapy, drug therapy, and surgery. So if a woman with symptoms visits a clinic, effective treatment is available in principle, although the likelihood of benefit varies between individuals.

Even so, the visit rate was low, and the reason seems to be cognitive rather than economic. Among women who were willing to pay, the median WTP per day was 100–200 yen depending on bother, while the actual out‐of‐pocket cost of a beta‐3 agonist widely used for OAB was estimated at about 14–42 yen per day. So even in the low‐bother group, the WTP was two to seven times higher than the actual cost. In other words, the treatment cost may be lower than patients think, but this information does not reach them. In fact, the most common reason for not visiting (n = 350) was “thought it was just aging” (65.1%), while “embarrassed” (13.1%) and “worried about cost” (13.4%), which we expected as female‐specific barriers, were not the main reasons. This dominance of “normalization of aging” is shared with other countries. In the midlife women cohort of Waetjen et al., the reasons for not seeking care were “not bad enough” (73%) and “a normal part of aging” (53%) [13], and a recent US study also reported that about half of women who delayed care thought it was “not treatable” [14]. In our study, 47.3% of symptomatic women answered that they “did not know” the symptoms can be treated, which agrees with these results. As for coping, 33.6% used over‐the‐counter drugs or pads, while only 4.5% received treatment at a medical institution. Self‐management with pads can keep the QOL for a while, but it may also fix a treatable condition as a “daily inconvenience to cope with” [15, 16]. This idea that symptoms are an “unavoidable part of aging,” together with the lack of awareness that treatment is possible, has been reported repeatedly in both Eastern and Western countries [13, 14, 15]. As also pointed out in JaCS 2023 [2], we need to teach society that OAB and urinary incontinence are treatable conditions, and that the cost is within the range that patients are willing to pay.

This study has several limitations. First, the web survey respondents may include patients who have already visited a clinic and received drugs or treatment effects, and we could not exclude them. Second, this is a cross‐sectional study, so we cannot infer a causal relationship between bother and care‐seeking. Third, because the respondents were registered members of an internet survey panel, the representativeness is limited. In addition, symptom bother was assessed with a single, nonvalidated numerical rating, and the None/Low/High grouping was a pragmatic classification without an established cut‐off; it was intended to indicate trends rather than to provide an absolute measure, and validated instruments (OABSS and ICIQ‐UI SF) were used in parallel.

5. Conclusion

In Japanese women aged 40 years or older, care‐seeking for LUTS was very low, and symptom bother and measured symptom severity were each independently associated with visiting a clinic. Because most women did not seek care mainly because they thought the symptoms were just aging, awareness that these symptoms are treatable should be promoted to close the treatment gap.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: Supporting Information.

Figure S1: Distribution of the symptom bother score and the None/Low/High cut‐offs. Symptom bother was rated on an 11‐point numerical rating scale (0–10). (A) All women (N = 1800); (B) symptomatic women (n = 402). Bars show the percentage of women at each score. A score of 0 (None) is shown in gray, 1–4 (Low) in teal, and 5–10 (High) in dark blue; the dashed lines mark the group boundaries. In the whole sample the score was strongly right‐skewed (63% scored 0), whereas among symptomatic women it was more symmetrically distributed. The cut‐offs were pragmatic and based on this distribution: 0 was the modal value and was treated as a separate group, and 5, the midpoint of the scale, was used as the lower bound of the high‐bother group.

Figure S2: Clinic‐visit rate by raw score with logistic fit. Clinic‐visit rate by each raw OABSS score (A) and ICIQ‐UI SF score (B). Dots are the observed rates, and the red line is the logistic fit. The confidence intervals are wide at high scores because only a few women had such high scores (e.g., OABSS ≥ 9 was only 2.3%). This is not because of the total sample size.

Table S1: Geographic distribution of respondents compared with the female population of Japan. Respondents were obtained from all 47 prefectures. The number and percentage of respondents in each prefecture are shown alongside the corresponding share of the 2024 female population of Japan. The distribution closely matched the national female population (Spearman's ρ = 0.89; Pearson's r = 0.98 for the percentage share; both p < 0.001). Population source: Statistics Bureau of Japan, Population Estimates, October 1, 2024.

Data Availability Statement

The de‐identified analysis dataset that supports the findings of this study is provided as a Supporting Information file (Data S1).