Abstract
This study examined treatment patterns and patient characteristics of men initiating alpha adrenergic blocker therapy (alpha-blocker) for benign prostatic hyperplasia (BPH). The 2009 Thomson Reuters MarketScan® Database was used to identify the newly initiated alpha-blocker: men ≥40 years old with continuous medical and pharmacy coverage for 12 months before and after alpha-blocker initiation, with no alpha-blocker or 5-alpha-reductase inhibitors in the previous year, and with ≥1 BPH diagnosis within 1 month before and 6 months after alpha-blocker initiation. This study analyzed patient demographics, clinical characteristics, adherence (percentage of men achieving medication possession ratio [MPR] ≥ 0.8), restarting the same alpha-blocker after discontinuation, switching to another BPH medication, and type of alpha-blocker (alpha 1 type selective or alpha 1 subtype selective agents). T tests and chi-square tests compared differences at the .05 significance level. A total of 13,474 men met the study criteria (mean age of 63.1 years). Two thirds of the men discontinued alpha-blocker in the 12-month period, among which restarts or switches were statistically different (p = .036) but numerically similar across cohorts. Adherence for alpha 1 type selective agents versus alpha 1 subtype selective agents at 6 months was 43.3% versus 38.1% (p < .01); at 12 months, 34.4% versus 30.5% (p < .01). Alpha-blocker discontinuation rates were high, which confirms low medication adherence reported among medications for several other chronic conditions; therefore, it is necessary to understand the reasons for alpha-blocker discontinuation.
Lower urinary tract symptoms (LUTS) suggestive of benign prostatic hyperplasia (BPH), hereafter referred to as BPH-LUTS (Wei, Calhoun, & Jacobsen, 2007), constitute bothersome chronic symptomatic conditions that are increasingly common as men age (Lepor, 2004; Roehrborn, 2008). By definition, BPH is a histological diagnosis that is made following prostate biopsy; however, BPH-LUTS is often identified and treated based solely on clinical presentation (Fourcade et al., 2012). The LUTS, suggestive of BPH, include symptoms such as urinary frequency, incomplete voiding, and nocturia. Symptoms may be measured by clinically validated instruments, such as the International Prostate Symptom Score (Barry et al., 1992) and may serve as a measure of symptom severity at baseline and a measure of efficacy following treatment.
One class of therapeutic agents for the first-line treatment of mildly to moderately bothersome BPH-LUTS is generally called “alpha-blockers,” and the physiologic action of this class blocks the alpha 1 adrenergic type receptors of the lower urinary tract. Among the alpha 1 types, there are three subtypes: alpha 1a/c, alpha 1b, and alpha1d (Schwinn & Roehrborn, 2008). Of the therapeutic agents, prazosin, doxazosin, and terazosin are alpha 1 blockers that block all three subtypes, and they are hereafter referred to as alpha 1 selective (Schwinn & Roehrborn, 2008). Tamsulosin, alfuzosin, and silodosin are considered alpha 1 subtype selective and are hereafter referred to as alpha 1 subtype selective (Table 1; Schwinn & Roehrborn, 2008).
Baseline Patient Characteristics for all Benign Prostatic Hyperplasia Patients.
Note. BPH = benign prostatic hyperplasia; COMP = comprehensive health plan (preventative and catastrophic coverage); HDDP/CDHP = high deductible or consumer-directed health plan; N/A = not applicable; SD = standard deviation. T test was used for continuous variables and chi-square test for categorical variables.
Depending on plasma concentration, alpha 1 selective agents (i.e., prazosin, doxazosin, and terazosin) may reduce arteriolar resistance and may be associated with various vascular sequelae, such as dizziness, lightheadedness, fainting, hypotension, and postural hypotension (Lepor, 2008). More recently developed alpha 1 subtype selective agents (i.e., tamsulosin, alfuzosin, and silodosin) have similar high affinity and selectivity for the alpha 1A adrenergic receptors that are located in the prostate and have a lower affinity for receptors that are located in the blood vessels. These agents have been reported to work effectively to relieve the symptoms of BPH-LUTS, with a low incidence of orthostatic and vasodilatory side effects (Lepor, 2008).
There are little data measuring alpha-blocker medication adherence, and there is little information about factors that might affect patient adherence to these therapies. Medication adherence across other chronic conditions is low, regardless of the severity of the disease or whether the disease is symptomatic (Yeaw, Benner, Walt, Sian, & Smith, 2009). The evidence suggests that adherence is not a drug-specific behavior, but understanding patterns of specific drugs or conditions may provide more tailored pathways to improve patient adherence and, ultimately, to patient health. In the case of BPH-LUTS, in which symptoms often interfere with functional status, adherence may improve symptom severity and health-related quality of life (Fourcade et al., 2012; Wilson & Cleary, 1995).
Given the relative scarcity of literature on adherence to alpha-blockers, this study uses a medical claims database to examine several key factors. The main objective of this study is to describe adherence patterns among men who initiate alpha-blocker therapy to treat BPH-LUTS and to evaluate treatment patterns that may influence adherence.
Method
Study Design, Sample Selection, and Data Source
A retrospective analysis, funded and analyzed by Eli Lilly and Company, was performed by use of the 2009 Thomson Reuters MarketScan® Database to identify BPH-LUTS patients who initiated alpha-blocker therapy in 2009. The date of alpha-blocker initiation was defined as the index date. Men were included in the study if they were 40 years or older at index with continuous medical and pharmacy coverage for the 12 months before and 12 months after alpha-blocker initiation (Figure 1). Men were excluded from the study if they had documented use of an alpha-blocker, 5-alpha-reductase inhibitor (5-ARI), or other BPH-LUTS therapy in the previous year, or if they had a BPH-LUTS-related procedure code 1 month before and up to 6 months after alpha-blocker initiation (Wei et al., 2007). Diagnostic and procedure codes were derived from the National Institutes of Health publication, Urologic Diseases in America, to provide a more objective definition of the disease (Wei et al., 2007).
Population attrition.
Statistical Analyses
Patient demographics, alpha-blocker persistence and adherence, other patterns of alpha-blocker use, and annual medication co-pay data were summarized for all BPH-LUTS patients and by the alpha 1 type selective (i.e., prazosin, doxazosin, and terazosin) or alpha 1 subtype selective (i.e., tamsulosin, alfuzosin, and silodosin) therapy cohorts. Persistence of alpha-blocker use was defined as the length of therapy from the index date to the earliest ending date of the last prescription, which was defined as either the date of the first gap of >60 days between prescriptions or the end of the follow-up period (6 or 12 months). Adherence to alpha-blockers was measured by medication possession ratio (MPR) in the 6-month and 12-month follow-up periods. The numerator in the MPR is the total daily supply of all of the prescriptions over the period, and the denominator is the duration from the index date to the end of the period. The overall adherence rate was defined as the percentage of patients with an MPR ≥0.8 (i.e., prescription coverage for 80% or more of days in the follow-up period). Patterns of alpha-blocker use that were examined included continuing or discontinuing therapy. Of those who discontinued therapy, patterns of restarting, switching, or terminating alpha-blocker therapy were evaluated. Continuation was defined as having no gap of greater than 60 days in prescriptions over the 1-year follow-up period. Discontinuation was defined by a gap of greater than 60 days in prescriptions within the 1-year follow-up period. Restarting therapy was defined as a prescription-fill within the year of the same alpha-blocker therapy, after a gap of greater than 60 days. Switching therapy was defined as a prescription-fill of a different alpha-blocker or a new BPH-LUTS medication, such as a 5-ARI, after discontinuing the initial alpha-blocker. Terminating therapy was defined as discontinuing the initial BPH-LUTS medication without restarting it or any other BPH-LUTS medication. T tests and chi-square tests compared differences at the .05 significance level. A sensitivity analysis was performed to assess whether the addition of a 5-ARI within 1-year postindex had any effect on adherence or persistence to alpha-blocker therapy, as the safety profile of a combination of alpha-blockers and 5-ARIs may not be well tolerated by patients.
Results
A total of 13,474 men, with a mean age of 63 years, met inclusion criteria for the study; of these men, 87.2% used an alpha 1 subtype selective agent (i.e., tamsulosin, alfuzosin, and silodosin; Table 1). Overall, 38.8% of men were adherent at 6 months, and 31.0% of men were adherent at 12 months (Figure 2). Among the men who discontinued therapy, 10.3% restarted on the same alpha-blocker therapy, 8.7% switched to a different BPH-LUTS therapy, and 47.1% discontinued entirely (terminated; Figure 3).
Alpha-blocker therapy adherence: total and cohorts.
Alpha-blocker therapy pattern for all BPH patients within 1 year.
There were few differences in baseline characteristics between the alpha 1 type selective (i.e., prazosin, doxazosin, and terazosin) and alpha 1 subtype selective (i.e., tamsulosin, alfuzosin, and silodosin) cohorts; due to the large sample size, they are statistically significantly different but not considered to be meaningful (Table 1). The average annual co-pay for alpha-blocker therapy was $96 (SD = $121), which constitutes about 18% of all prescription drug co-pays. Mean co-pays for alpha 1 subtype selective (i.e., tamsulosin, alfuzosin, and silodosin) users were significantly higher ($104 ± 126 vs. $43 ± 60), and alpha-blocker co-pays constituted a greater percentage of all pharmacy co-pays among alpha 1 subtype selective (i.e., tamsulosin, alfuzosin, and silodosin) users. Men on alpha 1 subtype selective agents (i.e., tamsulosin, alfuzosin, and silodosin) were significantly less adherent at 6 and 12 months (38.1% vs. 43.3%, p < .001; 30.5% vs. 34.4%, p = .001; Figure 2). Men on alpha 1 type selective agents (i.e., prazosin, doxazosin, and terazosin) were more likely to restart or switch therapy on discontinuation. A total of 16.9% of men on alpha 1 type selective agents (i.e., prazosin, doxazosin, and terazosin) restarted and 15.1% switched therapies, compared to 15.4% and 12.9%, respectively, of men on alpha 1 subtype selective agents (i.e., tamsulosin, alfuzosin, and silodosin; p = .036; Table 2). Men on alpha 1 type selective agents (i.e., prazosin, doxazosin, and terazosin) were also more persistent; at 6 months persistence was 117.3 days for alpha 1 type selective agents (i.e., prazosin, doxazosin, and terazosin) compared to 110.8 days for alpha 1 subtype selective agents (i.e., tamsulosin, alfuzosin, and silodosin; p < .001), and at 12 months, persistence was 188.0 days versus 175.9 days (p = .002), respectively (data not reported). Continuation rates across time are almost identical for alpha 1 type selective (i.e., prazosin, doxazosin, and terazosin) and alpha 1 subtype selective (i.e., tamsulosin, alfuzosin, and silodosin) agents.
Alpha-Blocker Therapy Pattern for Cohorts Within 1 Year.
Continued alpha-blocker medication in 1-year follow-up period; a gap of ≤60 days between alpha-blocker prescriptions was considered medication continuation.
Discontinued alpha-blocker medication in 1-year follow-up period; a gap of >60 days between alpha-blocker prescriptions was considered medication discontinuation.
Restarted the same initial alpha-blocker medication, after a gap of >60 days within 1-year follow-up period.
Switched to a new benign prostatic hyperplasia medication (including a different alpha-blocker) within 1-year follow-up period.
Terminated benign prostatic hyperplasia medication within 1-year follow-up period.
Analyses to examine if 5-ARI use affected adherence and persistence to alpha-blocker therapy identified that adherence to and persistence with alpha-blockers remained similar after removing 5-ARI patients from the analysis (data on file).
Discussion
Nearly 40% of men who initiated alpha-blocker therapy continued for 6 months, and about a third of men continued for 1 year, which is similar to the adherence rates seen in other chronic diseases (Yeaw et al., 2009). Among men who discontinued, the majority did not restart or switch therapy in the course of the study. Modest differences emerged when adherence patterns by selectivity of alpha-blockers were examined. Men initiating therapy on alpha 1 subtype selective agents (i.e., tamsulosin, alfuzosin, and silodosin) were significantly less likely to continue therapy at 6 months and 1 year; they were also significantly less likely to switch or restart therapy than were men who initiated therapy on alpha 1 type selective agents (i.e., prazosin, doxazosin, and terazosin). While these findings were statistically significant, qualitatively, the average man on an alpha 1 subtype selective agent (i.e., tamsulosin, alfuzosin, and silodosin) was slightly less likely to continue, switch, or restart alpha-blocker therapy than the average man on an alpha 1 type selective agent (i.e., prazosin, doxazosin, and terazosin).
The cost of therapy might be one possible explanation for the marginal differences this study observed in adherence and persistence by type of alpha-blocker therapy (Choudhry et al., 2011). In this study, men on alpha 1 subtype selective agents (i.e., tamsulosin, alfuzosin, and silodosin) faced slightly higher cost-sharing, both as an average value and as a percentage of total pharmacy cost-sharing. All alpha 1 type selective agents (i.e., prazosin, doxazosin, and terazosin) and some alpha 1 subtype selective agents (i.e., tamsulosin, alfuzosin, and silodosin) were generic at the time of this study, so the cost differential is not surprising. At the time of this study, most patients (87%) were taking an alpha 1 subtype selective agent (i.e., tamsulosin, alfuzosin, and silodosin), which was more expensive than the alpha 1 type selective (i.e., prazosin, doxazosin, and terazosin) alternatives. Several studies have demonstrated that the cost of therapy has a significant, but marginal, effect on adherence (Choudhry et al., 2011; Zhang et al., 2010).
This study conducted subanalyses on patients who initiated 5-ARIs during the course of the study and hypothesized that these men might experience more medication-related side effects and, thus, discontinue at a higher rate; however, no difference in the subanalysis was seen. In the literature, a number of other factors, which could not be captured in our database, have been proposed as contributing to nonadherence. Some of these factors include therapy-related factors (e.g., adverse events or inconvenience), patient-related factors (e.g., the degree of bother the condition imparts), disease-related factors (e.g., symptom severity), and health-system factors (e.g., health care provider expectation setting). The literature tends to suggest that a constellation of these factors, and the medication-related and health plan-related factors that were measured in this study, contribute to patient medication-related behaviors (Kripalani, Yao, & Haynes, 2007).
This study has several limitations. The quality of claims data is dependent on the accuracy of reporting and coding; however, standardized diagnostic and National Drug Codes minimize the variation in coding. Incident cases are difficult to identify, making assessments of treatment phases difficult; this study may have identified men who had already tried alpha-blocker therapy at least once. By including a lead-in period and by requiring continuous health care coverage, this challenge was minimized, but it still stands as a limitation. Information for BPH-LUTS, such as severity of disease, which may influence patient behavior or reasons for discontinuation, is not captured in claims data. Notably the claims database is not subject to recall bias, as when patients are surveyed about pharmaceutical use and treatment patterns. Results from this analysis apply only to the insured population in health plans that are comparable to those included in the database; however, the 2009 Thomson Reuters MarketScan Database contains a large, well-defined population, which has been used frequently in research studies like this one.
Consistent with the broader literature about patient adherence in chronic diseases, this study finds that the majority of men discontinue alpha-blocker therapy, although about 20% subsequently resume or switch alpha-blocker therapy. The type of therapy seemed marginally important in observed treatment patterns; men on alpha 1 type selective agents (i.e., prazosin, doxazosin, and terazosin) are slightly more adherent to therapy than are men on alpha 1 subtype selective agents (i.e., tamsulosin, alfuzosin, and silodosin), and they are more likely to subsequently restart or switch therapy. Cost may play a key role in type of therapy, as men who faced smaller cost-sharing amounts adhered better to therapy than those who faced larger cost-sharing amounts. Other unmeasured dimensions, such as treatment efficacy and treatment expectations, likely also influence patient behavior in alpha-blocker use. Additional research is needed to understand how the components of cost, efficacy, medication-related side effects, patient lifestyles, and other factors work together to influence alpha-blocker use. It is important to understand reasons for alpha-blocker discontinuation because men who discontinue therapy will likely continue to experience bothersome symptoms.
Footnotes
Acknowledgements
The authors would like to thank Teri Tucker, BA, of inVentiv Health Clinical, for her editorial contributions to this article, and Rebecca McCracken, MSPH, also of inVentiv Health Clinical, for her writing contributions to this article.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by Eli Lilly and Company.