Breast cancer survivors are at increased risk for urinary incontinence (UI) due to patient age and also possibly due to therapy-related decreases in estrogen. However, many breast cancer survivors with UI with may not seek care. An annual wellness visit (AWV) is one way for patients to proactively address their health and prevent functional decline, and they may be especially useful for conditions like UI where patients may not otherwise seek care for their symptoms. In the United States, patients with federal Medicare insurance coverage are encouraged to attend an annual primary care wellness visit aimed at identifying modifiable health risks and providing suggestions for behavioral changes. These visits involve mandatory questions about a patient’s health status but also serve as a time point where patients may bring up other concerns. UI screening is not required, but it is a quality measure. Our hypothesis is that breast cancer survivors attending an AWV are more likely to have documentation and treatment of UI than those without an AWV.

Using the Surveillance, Epidemiology, and End Results-Medicare linked cancer registry 2011-2014, we identified women with breast cancer (stage 0-3), enrolled in Medicare parts A, B, and D at least 1 year prior to their cancer diagnosis date, who were alive and still enrolled at least 2 years after their cancer diagnosis date. Women with an AWV in the year after cancer diagnosis were identified by one or more of the following CPT® codes: “Welcome to Medicare Visit” (G0402) and “Annual Wellness Visit” (G0438 and G0439) in the Medicare administrative (billing) data. Diagnosis of UI after cancer was defined as the presence of one or more ICD-9 or ICD-10 codes related to UI in Medicare data. These codes were identified based on prior research as well as input from Urology and Urogynecology authors. Pharmaceutical treatment for UI was defined as a prescription fill of any of following at least once: oxybutynin, tolterodine, fesoterodine, solifenacin, darifenacin, trospium, and mirabegron. Procedural treatment for UI was determined if one or more of the following CPT® codes were reported: periurethral bulking (51715), implantation of neurostimulator (64561), urethropexy (51990, 51840), synthetic or fascial sling (57288), and cystoscopy with chemodenervation of the bladder (52287). The primary outcome was the percentage of women in each group (AWV or no AWV) with the composite binary outcome of “UI patient” or not; “UI patient” was defined as women with documentation of a UI diagnosis and/or any type of treatment for urinary incontinence. We compared characteristics of patients with and without an AWV using Chi Square for categorical variables and Student’s t test for continuous variables. Multivariate analysis was then performed to examine whether AWV was associated with post-cancer documentation of UI diagnosis or treatment using logistic regression models adjusted for the following: age, race, ethnicity, comorbidity indices, cancer stage, receipt of chemotherapy, receipt of anti-estrogen therapy, quartile of census-tract level of income, recipient of low income subsidy, urban/rural distinction, and UI diagnoses and treatments reported prior to their cancer diagnosis. Multivariate analyses are reported as adjusted odds ratios (aOR) with a 95% confidence interval (CI).

A total of 12709 breast cancer survivors met inclusion criteria; 2433 (19.1%) women had attended an AWV and 10276 (80.9%) had not. There were statistically significant differences between the groups (Table 1). Women with an AWV were slightly younger, more likely to be white, married, have greater income, have fewer comorbidities, and have earlier breast cancer stages at diagnosis. Overall, 1510 (11.9%) of all patients met the criteria for UI patient status (Table 2). 288 (11.8%) of women with an AWV were UI patients compared to 1222 (11.9%) of women without an AWV, which was not different between groups (p=0.94). Reported UI-related diagnoses, pharmaceutical interventions, and procedural treatments were similar between cohorts. On multivariate analysis, women attending an AWV were still not more likely to be a UI patient (aOR 1.03, 95%CI 0.88-1.21). The most powerful predictor of being a UI patient in the logistic regression model was pre-cancer UI diagnosis or treatment (aOR 18.6, 95%CI 16.2-21.2). Given this strong association, we performed an ad hoc sub-analysis on the group of women with no prior UI diagnosis, procedure, or medication before their breast cancer diagnosis. Of these 11314 women, 2170 (19.2%) of them had attended an AWV and 9144 (80.8%) had not. 137 (6.3%) in women with an AWV and 594 (6.5%) in women without an AWV were UI patients; this was not different between groups (p=0.76). On multivariate analysis, this remained true; women attending an AWV were not more likely to be a UI patient (aOR 0.95, 95%CI 0.79-1.14).

In a large cohort of Medicare-insured female breast cancer survivors, an AWV within two years of diagnosis of breast cancer did not increase the incidence of UI diagnosis or treatment. This remained true in a cohort with no pre-cancer diagnosis of UI. Previous diagnosis and treatment of UI was the most powerful predictor of post-cancer UI diagnosis or treatment. The study was limited by an inability to adjust for primary care other than annual wellness visits.

Despite Medicare’s goals for annual wellness visits to identify modifiable health risks and provide behavioral and other interventions to improve overall health, these visits were not associated with improved urinary incontinence care delivery for a high-risk group (breast cancer survivors). Future studies should examine what interventions, including better focus on AWVs recognition of conditions like UI, could help maintain patients’ functional status and quality of life.

To Be Completed