Patient–Provider Discussions About Urinary Incontinence Among Older Women

Abstract

Background

Urinary incontinence (UI) is prevalent in women and has been associated with decreased quality of life and institutionalization. Despite this, and the fact that several treatment options exist, few women discuss UI with clinicians. The aim of this study was to examine the proportion of middle aged and older women with urinary incontinence who have discussed UI with clinicians, focusing on female health professionals as a way to examine this question outside of issues of health care access.

Methods

Data are from the Nurses Health Studies (NHS), two ongoing observational, prospective, cohort studies. The surveys collected detailed information about UI, including frequency, amount and type. Women were also asked if they had discussed UI with a clinician. We used multivariable-adjusted logistic regression to estimate odds ratios (OR) of participants reporting discussion about UI.

Results

94,692 women with UI aged 49–91 years old were included in this study. Of these, 34% reported that they had discussed their incontinence with a clinician. Women with daily UI had 4.4 times greater odds of discussing it with clinicians when compared to those with monthly UI (OR = 4.36, 95% confidence interval [CI] 4.06–4.69). When controlling for severity of symptoms, the oldest women, greater than eighty years, were 20% less likely to have discussed UI with their clinician, compared to the youngest women (OR = 0.81, 95% CI 0.73–0.89).

Conclusions

A minority of women with UI, even among health professionals, discuss their symptoms with clinicians. Oldest women were the least likely to discuss their UI with a provider.

Urinary incontinence (UI) occurs in up to 60% of women and is known to adversely impact quality of life, emotional well-being, relationships, and productivity (1–6). In middle-aged women, urinary incontinence has been found to be a predictor of worsened physical function scores later in life (7). In older women, urinary incontinence is associated with higher rates of hospitalization, institutionalization, and mortality (6,8). Moreover, older women with incontinence have increased frequency of incontinence, more severe incontinence, and worse self-reported health status than younger women with incontinence (9,10).

Despite the fact that there are many treatment options for women with incontinence, studies suggest that discussions with clinicians about urinary incontinence are infrequent. In previous work among older women, we have demonstrated that urinary incontinence severity and type are the two driving factors of whether or not women have outpatient visits for UI, even among a population of nurses (11). However, less is known about characteristics of women who discuss UI with their clinicians across the spectrum of ages at which UI is common. Herein, we specifically examine whether women’s age is an important determinant of whether she discusses UI with a clinician. As in our previous work, we focus on female nurses, to reduce issues of health care access and consciousness in understanding clinician discussions about UI. Further, we have previously found that rates of health care utilization and spending are fairly similar between the NHS population and similarly aged white women in the general population, especially in the nurses with a chronic health condition (12,13). We also test whether an array of demographic and clinical factors affect the odds of discussions of UI across a broad age group. This study adds to the prior literature by providing a robust sample size with a broad age range that is inclusive of the geriatric population, which is known to be significantly impacted by urinary incontinence.

Methods

Study Population

The Nurses’ Health Study (NHS I) and the Nurses’ Health Study II (NHS II) began in 1976 and 1989, when female registered nurses returned mailed questionnaires about their medical history and lifestyle. Description of the NHS and survey methods have been previously published (1,14). Since 1996 (NHS I) and 2001 (NHS II), the full-length questionnaires have included questions about urinary incontinence. The NHS are ongoing and the current follow-up rate in both cohorts is about 90%. The Institutional Review Board of Brigham and Women’s Hospital approved both the NHS I and II.

Urinary Incontinence

The NHS I and II ask women, “During the past 12 months, how often have you leaked or lost control of your urine?” Response choices were “never,” “ less than once/month,” “once/month,” “2–3 times/month,” “about once/week,” and “almost every day.”

The questionnaire also asks women reporting any UI to quantify their urine loss by asking: “When you lose your urine, how much usually leaks?” Response options are “a few drops,” “enough to wet your underwear,” “enough to wet your outer clothing,” and “enough to wet the floor.” A reliability study among a subgroup of these nurses demonstrated high reproducibility of responses to these questions (14,15).

We measured UI severity using the validated Sandvik severity index based on the two questions described above (14,16). We calculated severity by multiplying the reported frequency of UI by the amount of leakage (16). Women with a total score of 1 to 2 were classified as having mild UI severity, those with a score of 3 to 4 as having moderate UI severity, and those with a score of 6–12 as having severe UI (16).

We defined “stress UI” as women reporting leakage occurring with coughing, sneezing, lifting, laughing, or doing physical activity. We defined “urgency UI” as urine loss with a sudden and urgent feeling need to go to the bathroom. UI classifications were based on the participants’ reports of their dominant symptoms. Women who reported that stress and urgency symptoms occurred equally were defined as cases of “mixed UI.”

UI duration was calculated as the time from a woman’s first report of UI until 2012 (NHS) or 2013 (NHS II) in 2-year increments based on the interview frequency and categorized into less than 10 years versus 10 or more years.

Discussion of UI with Clinicians

The NHS I (2012) and NHS II (2013) surveys asked women for the first time: “Have you talked to your health care provider about leaking urine?” Response choices were “No I have not” and “Yes.”

Measurement of Risk Factors

The biennial NHS questionnaires include information on demographic, health, and lifestyle factors. In our multivariable-adjusted regression we include the following variables: age, race, employment status, smoking status, obesity, physical function, mental health score, postmenopausal hormone use, comorbidities, UI frequency, UI severity, UI type, and UI duration. We did not include other factors, such as beverage intake (eg, caffeinated beverages), which were not related to UI in our previous research (17).

Population for Analysis

For the current analysis, we included women with UI data on the 2012 (NHS I) or 2013 (NHS II) questionnaires with report of UI and who answered the question regarding if they talked with their health care provider about leaking urine. Women who reported not leaking urine were excluded.

Statistical Analysis

We used descriptive statistics to assess participants’ self-reported demographic and health characteristics (1). To test the relationship between respondent age and odds of provider discussions, we combined the NHS I and NHS II cohorts and used multivariable logistic regression adjusted for all covariates listed above to estimate odds ratios (OR) and 95% confidence intervals (CI). To examine each variable by age group, we constructed two separate models: one for NHS I (age range 66–93 years) and one for NHS II (age range 48–68 years), and performed multivariable logistic regression including all covariates described above. We used SAS 9.4 for statistical analysis (SAS Institute Inc., Cary, NC).

Results

Respondents were mostly Caucasian women with a mean age of 77 years in the NHS I cohort and 59 years in the NHS II cohort (range 48–93). 94,692 women in the NHS I and NHS II cohorts reported urinary incontinence. A minority of women in either cohort had talked to a clinician about their UI (38% NHS I vs 31% NHS II, p < .0001). However, UI symptoms were worse in the older women, with the NHS I group having more frequent and severe UI symptoms of longer duration (Table 1).

Table 1.

Baseline Characteristics for Women in the Nurses’ Health Studies with UI in 2012/2013 According to Whether Reported Talking with a Provider About Their UI

	Nurses’ Health Study II Women 48–68 y of age		Nurses’ Health Study I Women 66–93 y of age
	Discussed UI with Provider (n = 14,938)	Did not Discuss UI with Provider (n = 33,529)	Discussed UI with Provider (n = 17,593)	Did not Discuss UI with Provider (n = 28,632)
Demographic Factors
Mean age (SD), years	59.4 (4.6)	59.0 (4.6)	77.1 (6.6)	76.6 (6.6)
White race, %	97	97	98	98
Employment status, %
Not employed outside the home or retired	32	28	90	89
Employed outside the home-nursing	62	65	7	8
Employed outside the home-non-nursing	6	7	3	4
Health Behaviors
Smoking Status, %
Never	66	65	46	46
Past	30	30	51	50
Current	4	5	3	5
Health Status
Obesity (≥30 kg/m2), %	36	33	25	21
Comorbiditiesa, %
0	26	30	8	9
1	37	37	26	28
>1	37	33	67	63
Mean SF-36 Physical Function Score (SD)	-	-	59.8 (28.7)	65.1 (27.3)
Mean Mental Health Scoreb (SD)	6.7 (4.9)	6.3 (4.7)	2.0 (2.5)	1.8 (2.4)
Reported preventative health screening in the last 2 yc, %	99	98	97	96
Current postmenopausal hormone use, %	22	16	13	8
UI symptoms
UI frequency, %
<1/mo	23	51	13	41
1/mo	9	11	6	10
2–3/mo	18	17	15	17
1/wk	20	12	19	15
1/d	30	9	48	17
UI severity, %
Mild	34	65	20	53
Moderate	30	22	27	25
Severe	36	13	53	22
UI type, %
Stress	40	51	20	32
Urgency	26	22	37	35
Mixed	27	19	35	23
Other	6	8	8	10
Duration of UI,%
<10 years	12	20	7	15
10+ years	88	80	93	85

Note: Values of polytomous variables may not sum to 100% due to rounding. ^aComorbidities include high cholesterol, hypertension, type 2 diabetes, myocardial infarction, and stroke. ^bNHS II: Center for Epidemiologic Studies Depression Scale, score of 10 or more is cut point for depression, range: 0–30, NHS: Geriatric Depression Scale- score of 6 or more is cut point for depression, range: 0–15. ^cPreventative health screenings include physical exam, eye exam, mammogram or colonoscopy. UI = urinary incontinence.

Association Between Demographics, Health Behaviors, and Provider Discussions About UI

When combining both cohorts, we found that women older than 80 years were about 20% less likely than women below 55 years to discuss UI with a provider (OR 0.81, 95% CI 0.73–0.89; Figure 1). Women who reported greater health care use had higher odds of discussing UI with providers. For example, women who had preventative health screening were more likely to speak with providers about UI (NHS I OR = 1.5, 95% CI 1.34–1.69 and NHS II OR = 2.19, 95% CI 1.85–2.60; Table 2).

Figure 1.

Forest plot of adjusted odds ratio for discussing urinary incontinence (UI) with provide based on age group. Horizontal lines represent 95% confidence intervals (CI). Adjusted for demographics, health behaviors, health status and UI symptoms listed in Table 2.

Table 2.

OR^a and 95% CI for Report of Talking with a Provider About Their UI Among Women in the Nurses’ Health Studies with UI in 2012/2013 According to Demographics, Health Behaviors, Health Status, and UI Symptoms

	Nurses’ Health Study II Women 48–68 y of age		Nurses’ Health Study I Women 66–93 y of age
	Discussed UI with Provider (n = 14,938)	OR (95% CI)	Discussed UI with Provider (n = 17,593)	OR (95% CI)
Demographic Factors
White race, %	14,492	1.09 (0.96, 1.23)	17,249	0.92 (0.80, 1.07)
Employment status, %
Not employed outside the home or retired	4,369	1.00 (reference)	15,848	1.00 (reference)
Employed outside the home-nursing	7,938	0.82 (0.78, 0.86)	1,158	0.92 (0.84, 1.00)
Employed outside the home-non-nursing	754	0.82 (0.74, 0.91)	546	0.92 (0.82, 1.03)
Health Behaviors
Smoking Status
Never	9,763	1.00 (reference)	8,083	1.00 (reference)
Past	4,559	0.95 (0.91, 1.00)	8,930	0.99 (0.95, 1.03)
Current	579	0.68 (0.61, 0.76)	523	0.63 (0.56, 0.71)
Health Status
Obesity (≥30 kg/m2)	9,450	0.88 (0.84, 0.92)	4,309	0.98 (0.94, 1.01)
Comorbiditiesb
0	3,746	1.00 (reference)	1,314	1.00 (reference)
1	5,568	1.10 (1.05, 1.16)	4,494	1.06 (0.98, 1.15)
>1	5,642	1.10 (1.04, 1.17)	11,785	1.11 (1.03, 1.20)
Low SF-36 Physical Function Score	—	—	12,394	1.09 (1.04, 1.14)
Low Mental Health Scorec	3,422	0.98 (0.93, 1.04)	1,437	0.94 (0.87, 1.02)
Reported preventative health screening in the last 2 yd	14,745	2.19 (1.85, 2.60)	17,093	1.50 (1.34, 1.69)
Current postmenopausal hormone use	5,064	1.03 (0.99, 1.09)	2,612	1.84 (1.72, 1.98)
UI symptoms
UI frequency, %
<1/mo	3,406	1.00 (reference)	2,209	1.00 (reference)
1/mo	1,358	1.48 (1.36, 1.60)	976	1.57 (1.43, 1.73)
2–3/mo	2,724	1.94 (1.81, 2.09)	2,542	2.24 (2.07, 2.42)
1/wk	2,979	2.04 (1.83, 2.26)	3,391	2.23 (2.02, 2.47)
1/d	4,471	3.94 (3.55, 4.39)	8,475	4.78 (4.33, 5.28)
UI severity, %
Mild	4,960	1.00 (reference)	3,238	1.00 (reference)
Moderate	4,450	1.35 (1.25, 1.45)	4,322	1.26 (1.17, 1.37)
Severe	5,344	2.03 (1.83, 2.25)	8,761	1.97 (1.79, 2.17)
UI type, %
Stress	5,882	1.00 (reference)	3,373	1.00 (reference)
Urge	3,926	1.31 (1.24, 1.38)	6,387	1.32 (1.25, 1.39)
Mixed	4,069	1.29 (1.23, 1.37)	5,876	1.40 (1.32, 1.49)
Other	920	1.04 (0.95, 1.13)	1,346	1.07 (0.98, 1.16)
Duration of UI,%
<10 years	1,804	1.00 (reference)	1,230	1.00 (reference)
10+ years	13,134	1.41 (1.33, 1.50)	16,363	1.40 (1.30, 1.50)

Note: CI = confidence interval; OR = odds ratio; UI = urinary incontinence.

^aAdjusted for all variables included in the table and age. ^bComorbidities include high cholesterol, hypertension, type 2 diabetes, myocardial infarction, and stroke. ^cNHS II: Center for Epidemiologic Studies Depression scale, score of 10 or more is cut point for depression, range: 0–30, NHS I: Geriatric Depression Scale- score of 6 or more is cut point for depression, range: 0–15. ^dPreventative health screenings include physical exam, eye exam, mammogram, or colonoscopy.

Association Between Urinary Incontinence Symptoms and Provider Discussions About UI

UI frequency and severity were strongly associated with whether a woman had spoken to a provider about incontinence. As UI frequency increased, the adjusted odds of a woman discussing UI with providers increased (Table 2). For example, a woman with daily UI had four- to fivefold greater odds of talking to a clinician about incontinence than a woman with incontinence occurring once per month (NHS I: OR = 4.78, 95% CI 4.33–5.28; NHS II: 3.94, 95% CI 3.55–4.39; Table 2).

Women with severe UI (combining frequency and quantity of leakage) were twice as likely to discuss UI with their providers compared to women with mild UI (NHS I: OR = 1.97, 95% CI 1.79, 2.17; NHS II: OR = 2.03, 95% CI 1.83, 2.25; Table 2. UI type and duration were modestly related to provider discussions. Women with urgency (NHS I: OR = 1.32, 95% CI 1.25–1.39; NHS II: OR = 1.31, 95% CI 1.24–1.38) or mixed UI (NHS I: OR = 1.40, 95% CI 1.32–1.49; NHS II: OR = 1.29, 95% CI 1.23–1.37) were 30%–40% more likely to discuss UI with their providers than women with stress UI (Table 2). Additionally, the odds of women with UI for greater than 10 years to discuss UI were 40% greater than those with UI for less than 10 years (NHS I: OR = 1.40, 95% CI 1.30–1.50 and NHS II: OR = 1.41, 95% CI 1.33–1.50; Table 2).

Discussion

In this study, only 34% of nearly 95,000 women with UI reported having discussed their incontinence with a clinician. This is consistent with prior literature indicating that approximately 30% of women seek care for UI, despite our population including only women in the health care profession (4,5,18–34). After controlling for covariates, women more than 80 years were the least likely to have discussed UI with a provider compared to younger age groups, a potentially consequential finding given that UI is also associated with increased falls and nursing home placement. As expected, we find that characteristics of incontinence symptoms, especially the frequency and severity, had particularly strong relations to the odds of discussing UI with a clinician.

The evidence about the role of age in care seeking behavior for UI is contradictory (9,18,23,25–28,35). Our unique cohort provides a very large sample with data across middle aged through older women (48–93 years), thus providing unique advantages. We found that age was independently associated with decreased chance of discussing UI for the oldest women (aged > 80 years). We hypothesize that there are several factors that contribute to barriers for discussions about UI among women more than 80 years old. Prior studies have found that women’s attitudes towards aging, beliefs about the causes of incontinence, health expectations, and self-management and coping strategies are closely tied to whether they access the health care system to discuss urinary incontinence.” (36)

Our findings that the oldest women are the most likely to have incontinence symptoms and are also the least likely to have had a discussion with their provider about urinary incontinence has implications for patients, providers and policymakers. Urinary incontinence is associated with decreased health related quality of life in older women (10). Consistent with our findings, urinary incontinence has been referred to as the “neglected geriatric syndrome” on prior reports which surveyed the frequency which clinicians treated patients for this condition (37). Urinary incontinence is also associated with geriatric syndromes such as cognitive impairment and polypharmacy and is a leading cause for institutionalization among the older adults (3,10,38). Taken together, efforts to increase communication about urinary incontinence between women and providers could have significant impact on both patient well-being and health system costs.

In this study, we find that the characteristics of UI symptoms (UI type, frequency, severity, and duration) are directly associated with whether respondents reported talking to a clinician about their symptoms. This is consistent with prior literature about the association between patient discussions of UI and UI type (23,24,27,28,31) and symptom characteristics (18,21–27,29,31,32,39). This association reveals a missed opportunity to target women with milder UI symptoms, who may benefit from conservative, proactive interventions before their UI progresses and becomes more severe.

This study represents the largest yet to describe discussions about UI with clinicians among a broad age range of women. However, several limitations must be noted. First, the findings are based on the participants’ subjective reports of UI which may result in misclassification. Despite this, self-reported UI symptoms have been found to be reliable and valid compared to clinical assessment (40). Furthermore, the survey did not distinguish between patient and clinician-initiated discussions about UI. However, the factors we examined in our study may influence both patient and clinician-initiated discussions of UI. Finally, the NHS cohorts lack racial diversity which may limit generalizability. In addition, the NHS cohorts consist of nurses who may differ from the general population. However, interestingly, our data suggest that provider discussions among nurses are similar to other non-nursing cohorts, and we have also previously found that UI prevalence, incidence, progression and risk factors are similar between nurses and non-nurses (1,23,25). In fact, the finding that even nurses, who have higher health literacy than the general public, discuss incontinence infrequently makes clear how challenging increasing discussions will be. In spite of the limitations, this study sheds light on women who may be suffering with incontinence in silence. Identifying this silent population may help clinicians identify women who may benefit from UI treatments (41).

Conclusion

This cohort study of about 95,000 women with urinary incontinence finds that only 34% of these women have spoken to a clinician about their symptoms. Women who were greater than 80 years old had the smallest odds of having spoken to a clinician about urinary incontinence. These results may be a step towards identifying specific populations, such as women more than 80 years, who may benefit from initiatives to increase communication about urinary incontinence and treatment.

Funding

This study was supported by a grant from the National Institute of Diabetes and Digestive and Kidney Diseases (R01 DK105050). The Nurses’ Health Study is funded by grants from the National Cancer Institute (UM1 CA186107 and P01 CA87969 and U01 CA176726). The funding sources had no role in study design; in the collection, analysis, or interpretation of data; in the writing of the report; or in the decision to submit the article for publication.

Conflict of Interest

None reported.