Scoping review

A scoping review of the available research on the relationship and implications of UI in individuals with T2D was conducted. The scoping review method was selected to map the different types of evidence, summarize research findings, and identify gaps that warrant further research (6). The review will adhere to the methodology outlined in the Joanna Briggs Institute Reviewer's Manual (2) and is reported by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) (7). Although protocol registration in platforms such as PROSPERO is recommended to enhance transparency and avoid duplication, the present study did not meet PROSPERO’s eligibility criteria because scoping reviews are currently not accepted for registration. However, the full protocol was developed prior to conducting the review and is provided as supplementary material to ensure methodological transparency and reproducibility.

Research question

What is the extent of published evidence on the relationship between urinary incontinence and type 2 diabetes?

Research sub-questions

1. 1. What is the epidemiological association between type 2 diabetes and urinary incontinence?

2. 2. What is the clinical impact of urinary incontinence in people with type 2 diabetes?

3. 3. What are the available treatment options for urinary incontinence for people with type 2 diabetes?

Inclusion criteria

Studies were eligible if they reported research on T2D and UI. The following types of research were included: clinical trials and posterior analysis; epidemiological studies, including cohort and case-control studies; cross-sectional analysis; case series and case reports; and qualitative research studies. Poster publications, narrative reviews, and letters to the editor were excluded.

Search methods for identification of studies

Electronic searches

The following databases were searched from inception through December 2024: MEDLINE (Pubmed) and Embase. Studies in any language, from any country, and from any date were included. In addition, we screened reference lists from relevant published studies. The details of the full search strategy are provided in Supplementary Table 1.

Data collection and charting

Study selection

To identify the studies to be included, two authors (CEBM and AHRR) independently scanned the title and abstract of every record retrieved in the search. All potentially relevant articles were read in full. A third reviewer was available to resolve conflicts, but it was not required. The selection process followed the recommendations of the PRISMA-ScR checklist (2), and a PRISMA flowchart of study selection was provided following the PRISMA statement (8).

Data charting

Relevant information from each included study was extracted using a standardized data charting form that captured the following fields: author(s), year of publication, country of origin, type of research, type of population, age of the included population, and main findings related to the review. This approach allowed for a comprehensive summary of the critical conclusions.

Presentation of the results

The findings are presented in a narrative report that synthesizes the extracted data. The report is structured into sections corresponding to the research sub-questions. Qualitative studies were identified during the search, providing valuable insights not initially anticipated in the predefined sub-questions. Consequently, an additional section was incorporated to include these findings.

Nominal group technique

The project coordinating committee consisted of two authors (CEBM, AHRR). A focus group was conducted iteratively to plan the questions and statements for discussion in the Nominal Group Technique (NGT). The expert panel consisted of five specialists (9): two endocrinologists, one expert in domiciliary care, one geriatrician, and one registered nurse. Panelists were selected using a stakeholder analysis (10).

A modified NGT was used, retaining its main phases:

• (i) The Nominal or Silent Phase, where participants individually considered a question presented to them and wrote down their responses;

• (ii) The Item Generation Phase, where participants shared their responses with the group in turns, with items recorded without discussion;

• (iii) The Discussion and Clarification Phase, where group members discussed, clarified, and elaborated on the items, combining similar ones and removing duplicates;

• (iv) The Voting Phase, where participants ranked the items to establish priorities (9, 11).

The Nominal or Silent Phase was conducted using an online questionnaire. The Item Generation and the Discussion and Clarification Phases were held via videoconference, and the meeting was transcribed using the artificial intelligence tool Read AI. Finally, the Voting Phase was carried out through an online form, using a 9-point Likert scale for each idea.

The expert panel identified key barriers and challenges, which were organized and subsequently assessed by the panel members in terms of their relevance, potential impact, and feasibility of being addressed within the context of the Colombian healthcare system. Additionally, the experts developed a list of key healthcare team members whose involvement in caring for individuals with UI and T2D was considered essential. These aspects were evaluated using a 5-point Likert scale (1 = Very Low, 2 = Low, 3 = Moderate, 4 = High, 5 = Very High).

Panel recommendations

The online questionnaire responses were exported to a spreadsheet for consensus analysis. A threshold level of agreement of 70% among participants was established using the following ranges:1-3 points for limited importance, 4-6 for important but not critical, and 7-9 for critically important. In cases of strong disagreement, characterized by one panel member scoring a 1 and another scoring a 9, outliers were excluded, and the remaining responses were reviewed for consensus. For each statement, the median and interquartile range were calculated to illustrate variability, and all statements achieving consensus were reported (12).

Qualitative and quantitative data were collected during the NGT session and analyzed thematically to identify key themes. Since achieving consensus was not the primary objective of the NGT, most of the data are presented as thematic analysis results in the form of statements. Subsequently, a draft of the analysis was shared with the participants, who confirmed that the data accurately represented the discussions.

Scoping review

The database and resource searches identified 267 records. After duplicate removal, 190 records remained for screening, of which 60 were selected based on title and abstract evaluation. Following full-text assessment, 28 studies were excluded. The PRISMA flow diagram detailing the study selection process and reasons for exclusion is presented in Figure 1. The key characteristics of the included studies are summarized in Supplementary Table 2.

A total of 32 studies were analyzed to assess the association between T2D and UI (13-44), covering a wide range of study designs, populations, and methodologies. The studies were conducted across different countries, including the United States, Canada, China, Denmark, Norway, Palestine, Jordan, India, Taiwan, and several European nations, reflecting a global interest in this comorbidity. The study designs included cross-sectional studies (n=22), cohort studies (n=5), randomized controlled trials (n=3), and qualitative research (n=2). Most studies utilized population-based health surveys (NHANES, LOFUS, EPINCONT), clinical registries, or hospital-based data.

The sample sizes varied significantly across studies, ranging from 123 to 118,000 participants. Most studies focused on women (n=28), with only a few including men (n=5), emphasizing the higher prevalence of UI in females. Participants’ ages ranged from 18 to over 85 years, with mean ages typically between 55 and 75 years. While most studies focused on postmenopausal women, some reported findings in older adults receiving home care services (26,39) or frail elderly populations (32).

Epidemiology of urinary incontinence in people with type 2 diabetes

Studies investigating the incidence and prevalence of UI in individuals with T2D have reported disparate results, mainly due to differing study designs and patient populations. Jackson et al. (13) found that UI affected approximately 60% of postmenopausal women, highlighting this group's notable prevalence of UI. Danforth et al. (19) observed a weekly UI incidence of 8.7%

in women with diabetes, significantly higher compared to the 5.3% observed in women without diabetes, reinforcing the association between diabetes and greater UI risk. Similarly, Devore et al. (26) reported that around 48% of women with diabetes experienced UI, while Ebbesen et al. (17) found a prevalence of 39% compared with 26% among non-diabetic women. These and other findings are presented in Table 1, providing further comparative data on the higher prevalence of UI in individuals with diabetes.

The studies are consistent in revealing that diabetes significantly increases the risk for UI. For instance, Lewis et al. (14) reported an OR of 1.63 (95% CI: 1.28-2.09) for UI among women with diabetes using insulin, suggesting a notably higher risk of UI compared to non-insulin users, who had an OR of 1.20 (95% CI: 1.00-1.45). This association has been demonstrated in other studies (19,21), which reported an approximately 2.5- fold higher risk of UI in individuals with diabetes. More recent studies, such as those conducted by Fwu et al. (43) and Li et al. (44), corroborate these earlier findings, presenting similar evidence that diabetes, particularly in combination with other metabolic factors, substantially increases the odds of developing UI. These odds ratios and their corresponding 95% confidence intervals are further elaborated in Table 1, providing a detailed overview of the epidemiological evidence linking diabetes to elevated UI risk.

In terms of risk predictors, various factors have been identified as contributing to the development of UI in populations with diabetes, with advanced age, high BMI, history of vaginal delivery, and increased parity emerging as key contributors. For example, Li et al. (42) reported that age is a significant predictor, with each additional year of life contributing to a slight increase in the risk of around 8%, emphasizing the growing burden among older women with diabetes. Similarly, BMI was a decisive risk factor for UI and was associated with an increased risk of 11% for each unit increase in BMI (kg/m²). Parity also emerged as a significant determinant, with a history of vaginal delivery associated with a much higher likelihood of developing UI (OR ~3.147, 95% CI: 1.508-6.568), as shown by Li et al. (42). Moreover, Li et al. (44) demonstrated that higher insulin resistance, as assessed using the HOMA- IR index, was independently associated with both increased prevalence and greater severity of UI; women with higher levels of insulin resistance had 5.7-fold higher odds of developing severe UI (95% CI: 1.685-19.276). All these predictor risk factors, alongside their 95% confidence intervals are detailed in Table 1.

Contrasting the findings between women and men, the prevalence and severity of UI are consistently higher in women with T2D than in their male counterparts. Most female studies report UI prevalence rates ranging from around 40% to over 50%, whereas the study by Mahishale et al. (35) in Indian men documented a considerably lower prevalence of approximately 15.4%. In addition, the magnitude of the associations is more significant in women. For instance, Fwu et al. (43) reported an OR of 2.33 for any UI in women with diabetes, compared to an OR of 1.35 for men. Moreover, the subtypes of UI varied by sex, with stress and mixed UI predominating among women, whereas urgency UI was more common in men. These findings suggest the presence of sex-specific mechanisms and highlight the potential influence of confounding factors, such as prostate conditions in men.

Across the studies, various methods have been employed to evaluate UI in individuals with T2D. Many investigations have relied on self-reported questionnaires that capture the frequency and severity of UI symptoms. For instance, several studies-such as those by Danforth et al. (19) and Li et al. (42)- used structured survey instruments that asked participants about the frequency of urine leakage and the amount lost, with some studies further classifying UI into subtypes (stress, urgency, and mixed) based on specific symptom patterns. The International Consultation on Incontinence Questionnaire Short Form (ICIQ-SF) was frequently utilized due to its high validity, reliability, and sensitivity in assessing UI and its impact on quality of life. In addition, population-based surveys like the National Health and Nutrition Examination Survey (NHANES) provided standardized UI questions that allowed researchers to compare prevalence rates across diverse demographic groups. Other studies complemented questionnaire data with additional clinical evaluations and physical measurements. For example, some researchers performed urodynamic or pad-weighing tests to corroborate self-reported data. However, many studies preferred the non-invasive, survey-based approach to facilitate large-scale data collection.

Clinical impact of urinary incontinence in people with type 2 diabetes

The impact of UI in individuals with T2D extends beyond physical symptoms, affecting the quality of life, healthcare resource use, sleep, daily functioning, and emotional and social well-being. The reviewed studies highlight these multidimensional effects, emphasizing the broader implications of UI in diabetic populations.

UI has a profound impact on quality of life, with affected individuals reporting declines in physical, mental, and social well-being. Studies consistently show that UI leads to increased distress, reduced independence, and limitations in social interactions. Many perceive UI as an inevitable consequence of aging and diabetes, which contributes to low rates of healthcare- seeking behavior. Moreover, the severity of UI correlates with poorer self-reported health, higher rates of functional disability, and lower overall life satisfaction (28,30,31).

Regarding healthcare resource utilization, individuals with UI and diabetes experience a higher burden on healthcare systems due to frequent medical visits, emergency department admissions, and increased need for home care assistance. UI is a predictor of hospitalization and institutionalization in long-term care facilities, with people with diabetes being at exceptionally high risk. Patients with UI are more likely to require nursing support, assistive devices, and prescription medications, leading to increased healthcare costs and resource demands (28,31).

The impact on sleep is particularly notable in patients with UI, as frequent nocturnal voiding disrupts sleep patterns and contributes to poor sleep quality. Studies demonstrate that individuals with UI report increased rates of insomnia, excessive daytime sleepiness, and reduced sleep efficiency. Nocturia, a common symptom in patients with diabetes-related UI, further exacerbates sleep fragmentation and negatively affects cognitive function and mood (28,35).

UI significantly affects daily life activities, with individuals modifying their routines to accommodate frequent urination. Many patients limit their fluid intake, plan their outings based on restroom accessibility, and avoid social gatherings due to fear of leakage. UI-related mobility restrictions also contribute to reduced physical activity and increased risk of falls, particularly in older adults with diabetes. The interference of UI with work productivity and caregiving responsibilities further highlights its disabling nature in daily life (35,38).

The emotional and social impact of UI is substantial, with many patients experiencing embarrassment, anxiety, and depression. Women with diabetes and UI report higher levels of psychological distress, body image concerns, and feelings of shame. The stigma surrounding this condition often leads to social isolation, withdrawal from intimate relationships, and reduced participation in recreational activities. Studies suggest that UI management could significantly improve self-esteem, mental health, and overall social well-being (28,36).

Treatment of urinary incontinence in people with type 2 diabetes

Multiple studies have examined the impact of medications, glycemic control, weight loss, and specific treatments for UI in individuals with T2D, highlighting their role in modifying UI risk, severity, and response to intervention. The selected studies provide insights into how pharmacological and non-pharmacological strategies influence UI outcomes, with notable sex-specific differences.

Impact of diabetes medications: The relationship between diabetes medications and UI risk remains complex, with some treatments exacerbating symptoms while others offer potential benefits. For example, insulin therapy has been linked to higher UI prevalence, particularly in older adults, possibly due to its association with weight gain and nocturnal polyuria (34).

Impact of non-diabetes medications: In the only study identified evaluating a non- diabetes medication for urinary incontinence in women with type 2 diabetes, duloxetine reduced the frequency of incontinence episodes by 60.77% in the treatment group compared with 33.33% in the placebo group. However, this difference did not reach statistical significance (P=0.107), likely due to the small sample size of participants with T2D. Importantly, quality of life, as measured by the I-QOL score, improved significantly in the duloxetine group (P=0.037). The authors hypothesized that the reduced efficacy in women with T2D could be related to diabetic neuropathy affecting urethral sphincter innervation and responsiveness to treatment. Higher rates of adverse effects, including nausea, fatigue, and dry mouth, were also reported in women with T2D compared to the general population. Overall, while duloxetine did not achieve a statistically significant reduction in incontinence episodes in all patients with T2D, it may offer subjective benefits in quality of life for those who respond positively (18).

Impact of glycemic control: The relationship between glycemic control and UI remains a critical area of investigation. Poor glycemic control (HbA1c ≥ 8.5%) has been associated with increased UI severity, likely due to its impact on neuropathy, polyuria, and impaired detrusor function (34). In contrast, studies have shown that improved glycemic control may reduce UI symptoms, particularly in cases where polyuria and detrusor overactivity are major contributors. However, this effect may be sex-specific, as women with poorly controlled diabetes exhibit a stronger association with urgency and mixed UI compared to men, potentially reflecting sex differences in bladder physiology and the influence of estrogen on metabolic pathways.

Impact of weight loss: Weight loss has been identified as a key modifiable factor in reducing UI symptoms, particularly in overweight and obese individuals with diabetes. Interventional studies have demonstrated that a 5-10% reduction in body weight can significantly lower the incidence of stress UI (SUI) and urgency UI (UUI) (24). The Look AHEAD trial found that women who underwent intensive lifestyle intervention had a 42% lower incidence of UI than controls, emphasizing the benefits of structured weight loss programs in UI management (29). Interestingly, men with diabetes did not experience as pronounced a benefit from weight loss interventions, suggesting potential sex-specific differences in the pathophysiology of UI.

Specific UI treatments: Evidence directly addressing treatments for UI in individuals with T2D is limited. Available studies suggest that pelvic floor muscle training (PFMT), bladder training, and pharmacological therapies may be less effective in women with diabetes, potentially due to neuropathy and muscle dysfunction (30). Given this gap, most recommendations for specific UI management in T2D are extrapolated from the general population, underlining the urgent need for high-quality trials in diabetic cohorts.

Qualitative research on urinary incontinence in people with type 2 diabetes

The following findings are derived from qualitative research exploring the experiences, challenges, and strategies of UI management in older adults with T2D. This qualitative approach highlights the perspectives of patients, caregivers, and healthcare providers, offering valuable insights into the real-world implications of UI in home care settings (40,41).

UI is a prevalent issue among older adults with diabetes, with a reported 33.7% daily prevalence in home care settings. It is associated with falls, fractures, anxiety, depression, and a decline in quality of life, as well as an increased risk of premature institutionalization. Despite its high prevalence, UI remains underdiagnosed and undertreated. More than 40% of home care patients experience UI, yet few receive appropriate medical treatment. Instead, there is a widespread reliance on absorbent products, which, while providing practical support, do not address the underlying condition. The financial burden of continence products further limits access to effective treatments, particularly for individuals with limited resources.

One of the key challenges in managing UI among older adults with diabetes is the fragmented nature of healthcare services. Home care nurses play a critical role in identifying and managing UI. However, they face limited training, lack of access to specialists, and a system prioritizing task-based care over holistic management. Poor coordination between home care, primary care, and hospital services further impairs effective UI treatment. Additionally, UI in this population is influenced by hyperglycemia, polyuria, diabetic neuropathy, and detrusor dysfunction, all of which complicate management. Many diabetes-related medications, such as diuretics and antihyperglycemic agents, exacerbate UI symptoms, requiring individualized treatment approaches. Due to these challenges, nurses and caregivers frequently resort to absorbent products rather than evidence-based interventions, underscoring the need for improved training and resource allocation.

Current UI management strategies include assessments of hyperglycemia, polyuria, neuropathy, mobility limitations, and medication effects, as well as evaluations of home safety, restroom accessibility, and the availability of continence products. While pelvic floor muscle training (PFMT), bladder training, and dietary modifications aim to improve symptoms, absorbent products remain the primary coping mechanism. However, systemic barriers-including lack of integration between healthcare providers, limited electronic health record access, and difficulty obtaining specialist referrals-continue to hinder effective management. The reliance on absorbent products over medical interventions highlights the urgent need for a more structured, multidisciplinary approach to UI care in diabetic patients.

Caregivers play a central role in UI management, with 72% of patients relying on a spouse or family member for assistance. However, caregivers often experience burnout and financial strain, particularly given the high cost of continence products, which are not covered by government programs in many regions. Older adults with diabetes and UI also face the added complexity of managing multiple chronic conditions, making it difficult to balance treatments without exacerbating symptoms. Addressing these challenges requires policy changes aimed at expanding access to continence products, strengthening nurse training, improving healthcare integration, and establishing standardized UI management guidelines for diabetic patients. Such measures would not only enhance patients’ quality of life but also alleviate caregiver burden and reduce strain on healthcare systems.