Objective
The purpose of this study was to estimate the effect of insurance status on pelvic floor physical therapy (PFPT) nonparticipation for the treatment of urinary incontinence.
Study Design
A cross-sectional study of women referred to PFPT for urinary incontinence between January 2009 and June 2010 was conducted. A telephone questionnaire was administered. Multiple logistic regression was used to identify risk factors for nonparticipation.
Results
Thirty-three percent of women with private insurance and 17% with other insurance were PFPT nonparticipants. On multiple logistic regression, women with Medicare were more likely to participate in PFPT (odds ratio [OR], 0.12; 95% confidence interval [CI], 0.01–0.72). Risk factors for nonparticipation included insurance noncoverage (OR, 103.85; 95% CI, 6.21–infinity) and a negative perception regarding the benefit of PFPT (OR, 5.07; 95% CI, 2.16–12.49).
Conclusion
Among women who were referred to PFPT for urinary incontinence, insurance noncoverage and negative patient perception of efficacy were risk factors for nonparticipation, although having Medicare was protective. Improving patient education and insurance coverage for PFPT may increase usage.
Urinary incontinence is a common and debilitating condition, with prevalence rates that range from 10-40%. The emotional impact of incontinence can devastatingly impact a woman’s self-perception and result in embarrassment, social withdrawal, and depression among other consequences. The financial impact can be similarly overwhelming. The total direct costs that are associated with urinary incontinence exceeded $16 billion in 2001. As our society ages, the financial and emotional implications of this condition will continue to grow.
Pelvic floor physical therapy (PFPT) is an established treatment option for the improvement of all types of urinary incontinence. Given its proven efficacy, it is recommended by The Agency for Health Care Policy and Research that women should undergo a trial of conservative therapy before undertaking more invasive treatments that include surgery. However, for unclear reasons, many women who initially opt for a trial of PFPT never initiate treatment. The 2008 National Institutes of Health State-of-the-Science Statement on the prevention of fecal and urinary incontinence in adults highlighted that little information exists regarding barriers to participation in PFPT and specifically the role of insurance status.
The primary objective of this study was to estimate the effect of insurance status on PFPT nonparticipation for the treatment of female urinary incontinence. A secondary objective was to identify other risk factors that contribute to PFPT nonparticipation. Knowledge about these barriers may be used to improve patient counseling, inform public policy and educate insurers.
Materials and Methods
Institutional review board approval was obtained through Women and Infants Hospital of Rhode Island. We performed a cross-sectional study of all women who opted for and accepted referral to PFPT for the treatment of stress, urge, or mixed urinary incontinence from January 2009 through June 2010. All women were recruited from our institution, an academic tertiary urogynecology center composed of 5 fellowship-trained urogynecologists. Our practice cares for women regardless of ability to pay and provides for an economically diverse population that includes uninsured, subsidized, and privately insured women. Women are counseled routinely regarding PFPT as a treatment option for urinary incontinence and are referred regardless of insurance status or knowledge of specific coverage issues. At the time of referral, women are advised to call their insurance plan to determine coverage of PFPT services.
Our division has a colocation model with our hospital-based PFPT program, which includes 3 experienced pelvic floor physical therapists with a clinical office in the same building as the urogynecology outpatient office. This provides streamlined patient care, access to medical records, and follow-up evaluation between the practices. All therapists have specialized training in female pelvic floor disorders. Once formal referrals are made to the PFPT program, women are contacted by the physical therapists to schedule initial evaluation and follow-up appointments. For patient convenience, our PFPT program offers daytime and evening appointments. Physical therapy (PT) regimens are tailored to the patients’ needs and typically consist of weekly appointments for 6 weeks then monthly follow-up evaluation as needed.
For our study population, we included only women who opted for a trial of PFPT for urinary incontinence after comprehensive counseling of all treatment options. We identified eligible women by reviewing formal referral requests to the PFPT program from our division during the study period. Eligible women were mailed a study packet that included a letter of introduction, a verbal authorization form, and a stamped, addressed “do not contact” postcard. Women who did not return the postcard were contacted by phone 2-4 weeks after the study packet was mailed. Women were excluded from our study if they were non-English speaking, were ≤18 years old, were referred for conditions other than urinary incontinence, had physical or cognitive impairments that would limit full engagement in PFPT, or declined participation.
After informed consent had been obtained, the hospital record was abstracted, and a 15-item telephone questionnaire was administered to identify potential barriers to PFPT participation ( Appendix ). Our chart abstraction and questionnaire were based on a conceptual framework developed by health services researchers to define access barriers to general PT services. This framework is composed of 3 broad categories that include (1) structural variables (variables that impact access such as age, insurance status, education, and income), (2) utilization variables (that refer to the availability of PT providers and PT referrals and an understanding of the role of the physical therapist), and (3) outcome measures that indicate access (defined by adherence to prescribed treatments, health behaviors, and health status).
We were primarily interested in evaluating structural variables as potential barriers to PFPT, and we included questions about insurance type and coverage for PFPT services as well as age, race, education, income, self-rated health status, medical history, and mobility. We confirmed incontinence type with the hospital record and assessed incontinence severity. We surveyed the distance that women would be required to travel from home to PFPT and the impact that travel would have on their ability to make an appointment for these services. With regard to utilization variables, we evaluated patient attitudes about PFPT efficacy with the questions “Do you think PFPT will help improve your symptoms of incontinence” and “Are you nervous about PFPT.”
Demographic and clinical information was collected through chart review. Medical history, comorbidities, physical examination, pelvic organ prolapse quantification measurements, and urodynamic data were collected. Women also routinely completed the short forms of the Pelvic Floor Distress Inventory and Pelvic Floor Impact Questionnaire at baseline.
The exposure was defined as insurance type that was confirmed by a review of the patient’s hospital record and categorized as private, compared with “other.” Other insurance was defined as public, subsidized, or uninsured. Our outcome of PFPT nonparticipation was defined as women who opted for PFPT but subsequently did not complete at least an initial evaluation with the pelvic floor physical therapist. Our outcome was based on patient report and confirmed with the use of both PFPT records and our institutional tracking system.
Descriptive statistics, the Student t tests and the Wilcoxon rank-sum and Fisher’s exact tests were performed as appropriate. Two-tailed probability values were calculated; a probability of < .05 was considered to be statistically significant. We used multiple logistic regression to identify risk factors for PFPT nonparticipation, with adjusted odds ratios and 95% confidence intervals. The linearity assumption was confirmed for all continuous variables. All variables that were associated with PFPT nonparticipation in the univariate models with a probability value of ≤ .1 were evaluated as potential confounders in the multivariate analysis; those variables that affected the model significantly were retained in the final model. Model discrimination was assessed by the C-statistic, and model calibration was tested by the Hosmer-Lemeshow goodness-of-fit test, with a probability value of > .05 indicating adequate fit. In addition, we used Firth’s penalized likelihood that was based on Allison’s report to evaluate the impact of variables that had a strong effect on the outcome (specifically no insurance coverage) but that resulted in convergence failures when conventional binary logistic regression was used because of small cell numbers and quasi-complete separation. All statistical analysis was performed with SAS software (version 9.1; SAS Institute, Cary, NC).
We were particularly interested in evaluating the role of insurance status on PFPT nonparticipation. There are limited data to guide expectations and assumptions regarding insurance status and PFPT nonparticipation; the plethora of insurance plans and options further complicates assumptions. Recognizing this, we assumed that 20% of women with private insurance and 50% of women with other insurance would be PFPT nonparticipants. With these assumptions, 30 women in each group would provide 80% power at an alpha of 0.05. Approximately 160 women with urinary incontinence are referred over a 12-month period to PFPT from our Division and approximately 80% had private insurance. Therefore, we estimated that recruitment over a 17-month time period would allow us to meet our target, accounting for loss to follow-up and nonparticipation in the study.
Results
A total of 213 women were referred to PFPT for urinary incontinence during the study period. Three women were excluded because they were non-English speaking, and 1 woman was ≤18 years old. Forty-eight women returned the “do not contact” postcard, and 15 women could not be reached by phone. One hundred forty-six women met the inclusion criteria and responded to the telephone questionnaire, for a response rate of 68.5%.
The mean age of our study population was 55.1 ±15.3 years; most of the women were white (86.5%). Approximately 78.1% of the women reported at least some college education, and 51.5% reported a yearly household income of ≥$60,000.
Thirty-three percent of the women with private insurance and 16.7% of the women with other insurance were PFPT nonparticipants ( Table 1 ). The women with private insurance were younger, more educated, had higher household incomes, and reported better self-rated health status compared with women with other insurance.
| Characteristic | Insurance, n (%) | P value | |
|---|---|---|---|
| Private (n = 110) | Other a (n = 36) | ||
| Physical therapy participation | .09 | ||
| Yes | 74 (67.3) | 30 (83.3) | |
| No | 36 (32.7) | 6 (16.7) | |
| Age, y | < .0001 (overall) | ||
| 18-29 | 10 (9.1) | 1 (2.8) | 1.0 |
| 30-44 | 23 (20.9) | 2 (5.6) | .7 |
| 45-64 | 66 (60.0) | 10 (27.8) | Reference |
| ≥65 | 11 (10.0) | 23 (63.9) | < .0001 |
| Education | .02 (overall) | ||
| High school or less | 18 (16.4) | 14 (38.9) | Reference |
| Some college | 36 (32.7) | 11 (30.6) | .08 |
| 4-y college graduate | 56 (50.9) | 11 (30.6) | .006 |
| Household income per year, $ b | .0005 (overall) | ||
| <20,000 | 13 (12.6) | 10 (32.3) | Reference |
| 20,001-40,000 | 12 (11.7) | 9 (29.0) | 1.0 |
| 40,001-60,000 | 16 (15.5) | 5 (16.1) | .2 |
| >60,000 | 62 (60.2) | 7 (22.6) | .001 |
| General health c | < .0001 (overall) | ||
| Excellent/very good | 74 (67.9) | 10 (27.8) | Reference |
| Good | 24 (22.0) | 17 (47.2) | .0004 |
| Fair/poor | 11 (10.1) | 9 (25.0) | .002 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
