David K. Blough, PhD

 To examine variables influencing colony-stimulating factor (CSF) prescription as primary prophylaxis versus other use during patients' initial chemotherapy course among a large sample of health insurance records.

 Adults 25 years or older with a diagnosis of breast, colorectal, or non-small cell lung cancer (NSCLC) between January 1, 2002, and December 31, 2005, were identified from the western Washington State Surveillance, Epidemiology, and End Results Seattle Puget Sound registry. We linked these records to health insurance claims. Chemotherapy regimens identified from insurance claims were categorized as carrying high, intermediate, or low risk of myelosuppression according to the National Comprehensive Cancer Network guidelines and the literature. Colony-stimulating factor use was described as primary prophylaxis, other use, or no use, and logistic regression analysis identified factors associated with CSF use.

 For patients with breast cancer, colorectal cancer, and NSCLC, respectively, 58%, 0%, and 28% received CSFs as primary prophylaxis in conjunction with high-risk chemotherapy regimens, whereas 10%, 7%, and 21% did so in conjunction with low-risk chemotherapy regimens. Prophylactic CSF use increased from 2002 to 2005 for breast cancer but remained constant for colorectal cancer and for NSCLC.

As primary prophylaxis, CSF use is underutilized based on recommendations for patients having cancer who receive chemotherapy regimens carrying high febrile neutropenia risk and may be overutilized for patients who receive chemotherapy regimens carrying low febrile neutropenia risk. Further research is needed to understand the barriers to implementing guidelines in clinical practice.

Based on practice guidelines, colony-stimulating factors (CSFs) are underutilized for primary prophylaxis and are overprescribed in settings where they offer little or no clinical benefit.

Prophylactic CSF use has increased among patients with breast cancer but has remained constant among patients with colorectal and lung cancer.

On average, less than 50% of patients with breast, colorectal, or non-small cell lung cancer received CSFs as primary prophylaxis in conjunction with chemotherapy regimens carrying high risk for febrile neutropenia.

Up to 21% of patients received CSFs alongside chemotherapy regimens that pose little or no risk for febrile neutropenia.

Febrile neutropenia (FN) is a common and serious complication of cytotoxic chemotherapy. Clinical practice guidelines

 recommend the use of colony-stimulating factors (CSFs) in  the first chemotherapy cycle as primary prophylaxis to lessen chemotherapy- induced FN incidence when the expected risk of FN is approximately 20% or higher.

 Individuals who experience FN are at risk for hospitalization

In clinical trials, prophylactic treatment of patients with CSFs at the start of the initial chemotherapy cycle can reduce FN incidence by as much as 50%.

 National Comprehensive Cancer Network (NCCN),

 and European Organisation for Research and Treatment of Cancer

 guidelines recommend the use of CSFs as primary prophylaxis when the anticipated incidence of FN associated with a chemotherapy regimen is approximately 20% or higher (high risk) or 10% to 20% (intermediate  risk) if patient risk factors could contribute to greater risk of FN. Routine CSF use is not recommended when the risk of FN associated with chemotherapy is less than 10%, nor is it recommended to treat FN or severe neutropenia.

Although these guidelines have existed for years, surveys suggest that physicians commonly prescribe CSFs for indications that are not supported by evidence or guidelines.

 No study to date has characterized the spectrum of CSF use among large populations of patients with cancer treated in community practice settings.

Using claims data from 4 large health insurance plans linked to cancer registry data, we evaluated patterns of CSF use among patients with breast cancer, colorectal cancer, or non—small cell lung cancer (NSCLC) who received myelosuppressive chemotherapy. We examined variables influencing use proximal to the initiation of chemotherapy to identify predictors of CSF prescription as primary prophylaxis versus other use.

Patient-level data obtained from the western Washington State Surveillance, Epidemiology, and End Results (SEER) Seattle—Puget Sound registry were merged with healthcare claims data from 4 major insurers. These included Medicare, Medicaid, Premera Blue Cross, and Regence Blue Shield.

The SEER Seattle—Puget Sound registry was established in 1974 under contract with the federal SEER program. The registry provides data on the incidence, tumor characteristics, diagnosis stage, initial treatment, and survival of all newly diagnosed cancers (except nonmelanoma skin cancers) occurring in residents of 13 counties in western Washington State.

Premera Blue Cross and Regence Blue Shield are 2 of the largest private health insurers in Washington State. Both are nonprofit, and Premera Blue Cross serves more than 1.4 million members in Washington State,

while Regence Blue Shield provides coverage to more than 1 million Washington State residents.19

The Medicaid program provides health insurance for approximately 420,000 low-income residents in Washington State. The program operates under the auspices of the  Washington State Department of Social and Health Services and the Health Recovery Services Administration.

The Medicare program,21 administered by the Centers for Medicare and Medicaid Services, provides coverage for persons 65 years or older, persons younger than 65 years with certain disabilities, and persons of all ages with end-stage renal disease. Medicare is composed of Part A (inpatient hospital care) and Part B (outpatient care and physician services).

The 4 health insurance databases used for this study include approximately 3.7 million Washington State residents. These residents represented more than 75% of the state’s population in 2005.

The study was approved by the appropriate institutional review boards. A waiver of consent was obtained for access to each database before linkage was performed.

To identify subjects with incident cancers in the 4 health plans, we cross-linked person-level identifiers (ie, full name, sex, date of birth, zip code) from each plan’s enrollment files with cancer cases identified in the SEER Seattle—Puget Sound registry. Other inclusion criteria were age 25 years or older at the time of diagnosis, at least 1 health insurance claim for cancer chemotherapy, and diagnosis date between January 1, 2002, and December 31, 2005. Patients were excluded if they had any other malignant neoplasm previously recorded in SEER (including a diagnosis of breast cancer, colorectal cancer, or NSCLC before January 1, 2002) or had incomplete claims records over the period of interest. We excluded male patients with breast cancer from the study. Claims were collected until 12 months after diagnosis or until date of death.

Chemotherapy Use, CSF Use, and FN Event Identification

The study focus was to determine CSF use in primary prophylaxis versus other settings relative to the FN risk of each patient’s initial chemotherapy regimen. Colony-stimulating factors were defined as filgrastim or pegfilgrastim. Although sargramostim may be used in some settings for chemotherapyinduced neutropenia,

it was excluded from analysis, as this use is not approved by the US Food and Drug Administration. We defined CSF administration as primary prophylaxis or as other use based on the date of CSF administration relative to the first date of chemotherapy administration.

When used as primary prophylaxis, CSFs should be administered 24 to 72 hours after chemotherapy administration.

 We extended this time range to account for potential coding misspecifications of administration date and to consider the common practice of giving patients CSFs to take home and self-administer following chemotherapy initiation. Therefore, we defined primary prophylaxis as CSF use at the first chemotherapy treatment, specifically 1 day before through 7 days after the first chemotherapy administration date. Other CSF use was defined as any use in other settings (eg, following an FN event).

Using insurance claims, we identified the initial chemotherapy agents administered, the dates each agent was started and stopped, the initial CSF use date, and FN events. Patients were grouped into categories based on the myelosuppressive risk of the chemotherapy regimen. We developed an algorithm to identify commonly used chemotherapy regimens. The algorithm, summarized in 

), considers the chemotherapy regimen administered and the timing relative to the diagnosis dateand to other prescribed agents. The myelosuppressive risk of each chemotherapy regimen (the risk of developing FN in the absence of CSF administration) was assigned based on the high-, intermediate-, and low-risk categories published by the NCCN23 during the same period as the study population’s diagnoses (2001-2005). In cases where a regimen was not categorized by the NCCN, a board-certified oncology phar-macist (JSM) assigned the regimen to a risk category based on FN incidence descriptions from the published literature. The chemotherapy risk categorizations are listed in 

Claims records were used to determine chemotherapy onset. To allow for lags in claims recording, records were searched from the initial administration date of the first chemotherapy agent to 14 days past the last date when a chemotherapy agent from the initial regimen appeared in the patient’s claims. If a different chemotherapy agent appeared during the 14-day window, then a new chemotherapy regimen was considered to have been initiated if at least 3 claims of the original chemotherapy agent had appeared or 2 expected cyclical time gaps had occurred since the first administration of the initial chemotherapy regimen (eAppendix A).

We used logistic regression analysis to identify factors associated with CSF use, which was categorized as primary prophylaxis, other use, or no use. The cohort was stratified into groups of high, intermediate, and low risk of FN before running the regression analysis. Independent variables included age, sex (for colorectal cancer and NSCLC), race/ethnicity, health insurance plan, cancer site, stage at diagnosis, noncancer comorbidity in the 12 months before diagnosis, and receipt of surgery or radiation therapy within 30 days of initial chemotherapy administration. Health insurance plan was grouped as Medicare, Medicaid, or commercial (Regence Blue Shield and Premera Blue Cross combined). Persons enrolled in more than 1 health plan were assigned to the plan containing most claims over the period of observation. We included all claims from both plans for dually enrolled persons.

Among 8354 patients with breast cancer, colorectal cancer, or NSCLC, 2728 patients were identified as having received chemotherapy during the study period. 

lists the characteristics of these individuals. Most initial chemotherapy regimens administered to patients with breast cancer were categorized as high FN risk (74% across health plans), whereas most chemotherapy regimens administered to patients with NSCLC and colorectal cancer were characterized as low FN risk (59% and 77%, respectively). Patients with NSCLC were older and had the highest proportion of individuals diagnosed at the most advanced stage (62% across health plans), whereas most patients with breast and colorectal cancer were diagnosed at the regional stage (56% and 59%, respectively). Patients with NSCLC had the highest mean comorbidity ratings among the 3 cancer groups.

Comparing health plans, a higher proportion of Medicaid patients were diagnosed as having distant-stage disease than those enrolled in Medicare or with the commercial insurers. The proportion of nonwhite patients with cancer enrolled in Medicaid was 21% to 24% vs 5% to 11% in the other health plans.

Across all tumor types, the proportions of all CSFs given as primary prophylaxis to persons with high-, intermediate-, and low-risk chemotherapy were 54%, 24%, and 15%, respectively. The proportions of CSFs given as primary prophylaxis in conjunction with high-risk chemotherapy were 58% for breast cancer and 28% for NSCLC. No patients with colorectal cancer received CSFs as primary prophylaxis in conjunction with high-risk chemotherapy. The proportions of patients with breast, NSCLC, and colorectal cancer receiving primary prophylactic CSFs and intermediate-risk chemotherapy were 46%, 24%, and 8%, respectively (

). In the low-risk chemotherapy setting, 10%, 7%, and 21% of breast, colorectal, and NSCLC patients received CSF as primary prophylaxis. Among patients with breast cancer, NSCLC, and colorectal cancer, the proportions receiving CSFs in any setting were 34%, 33%, and 21%, respectively.

The proportion of all patients with breast cancer receiving CSFs as primary prophylaxis increased from 9% in 2002 to 46% in 2005. Primary prophylactic CSF use was highest among patients receiving high-risk chemotherapy regimens and was lowest among patients receiving low-risk chemotherapy regimens (Table 2 and the 

receiving chemotherapy and any CSFs increased from 40% in 2002 to 74% in 2005. The increase was greatest among patients who were commercially insured (from 44% in 2002 to 80% in 2005) and was lower among patients who were enrolled in Medicaid (from 38% in 2002 to 61% in 2005) or in Medicare (from 34% in 2002 to 67% in 2005).

Among patients with NSCLC who received any CSFs, primary prophylactic use ranged between 20% and 26% from 2002 to 2005. The proportion of patients with NSCLC receiving chemotherapy and any CSFs increased from 32% in 2002 to 44% in 2005. The percentage increase was greatest among patients who were enrolled in Medicaid (from 27% in 2002 to 41% in 2005) or in Medicare (from 32% in 2002 to 45% in 2005) and was lowest among patients who were commercially insured (from 36% in 2002 to 39% in 2005).

Among patients with colorectal cancer who received any CSFs, 7% received primary prophylactic CSFs. Of those receiving high-risk chemotherapy regimens, 38% received any CSFs during chemotherapy, whereas 43% of patients receiving intermediate-risk chemotherapy regimens and 21% of patients receiving low-risk chemotherapy regimens received any CSFs. The proportion of patients with colorectal cancer undergoing chemotherapy who received any CSFs increased from 21% in 2002 to 25% in 2005. The increase was greatest among patients enrolled in Medicaid (from 13% in 2002 to 25% in 2005) or in Medicare (from 19% in 2002 to 25% in 2005). The percentage of commercially insured patients receiving CSFs decreased slightly (from 26% in 2002 to 24% in 2005).

Colorectal cancer was removed from the multivariate analysis for high-risk chemotherapy regimens because there was no primary prophylactic CSF use in this category. In addition, because of low proportions of minority races/ethnicities,the racial/ethnic categories were collapsed to white versus nonwhite. Multivariate analysis revealed that factors predicting CSF use as primary prophylaxis differed across tumor types (eAppendices A, B, and C). Among patients receiving high-risk chemotherapy regimens, patients with breast cancer were significantly more likely to receive CSFs as primary prophylaxis than patients with NSCLC (odds ratio [OR], 2.43; 95% confidence interval [CI], 1.07-5.49) (

). Among patients undergoing intermediate-risk chemotherapy regimens, patients with breast cancer were significantly more likely to receive CSFs as primary prophylaxis than patients with NSCLC (OR, 4.01; 95% CI, 1.06-15.15), and patients with colorectal cancer were significantly less likely to receive CSFs as primary prophylaxis than patients with NSCLC (OR, 0.24; 95% CI, 0.07-0.80) (

). Among patients undergoing low-risk chemotherapy regimens, patients with breast cancer (OR, 0.32; 95% CI, 0.11-0.93) and colorectal cancer (OR, 0.26; 95% CI, 0.11-0.66) were less likely to receive CSFs as primary prophylaxis than patients with NSCLC (

Using cancer registry and claims records, we examined patterns of CSF use among patients enrolled in 4 health insurance plans in Washington State receiving initial chemotherapy for breast cancer, colorectal cancer, or NSCLC. Although overall CSF use increased substantially over time, its use as primary prophylaxis was variable among cancer sites. Colony-stimulating factors were used as prophylaxis frequently for patients with breast cancer but in few patients with colorectal cancer or NSCLC.

Overall CSF use increased substantially from 2002 to 2005. The greatest increase occurred in women with breast cancer, such that in the last year of the observation period almost 75% received CSFs during their initial chemotherapy regimen. However, less than half of the CSF use was as primary prophylaxis. These results suggest that (with the possible exception of breast cancer) past and current guidelines addressing CSF use for patients receiving myelosuppressive chemotherapy have been ineffective in aligning actual practice patterns with use best supported by evidence and package labeling.

Multivariate analysis showed that for women with breast cancer the likelihood of primary prophylactic CSF use was predicted by administration of chemotherapy regimens with higher FN risk. Most clinical evidence supports CSF use in this setting, and guidelines have been in place for the longest time. Among patients with breast cancer administered high-risk chemotherapy herein, 28% received CSFs as primary prophylaxis, and 26% never received CSFs.

Among patients with NSCLC and colorectal cancer who received high- or intermediate-risk chemotherapy, prophylactic CSF use was low and did not change over time. Most of these patients were placed on regimens with an intermediate FN risk, in which guidelines recommend that consideration be given to CSF prophylaxis if other risk factors exist (eg, preexisting neutropenia or poor performance status).

 Patients with NSCLC and colorectal cancer were older and had more noncancer comorbidities than patients with breast cancer, which may have motivated physicians to prescribe lower-intensity chemotherapy doses, thereby lowering the perceived need for prophylactic CSFs. Unfortunately, information for evaluating chemotherapy doses delivered was unavailable among the claims data used for this study.

One clinical trial evaluated the efficacy of CSFs among patients with small cell lung cancer

; there are no direct studies of CSF use among patients with NSCLC. Some controversy persists as to the relative effectiveness of CSFs among speciÃ¯Â¬Â

 From the Fred Hutchinson Cancer Research Center (SDR, CLM), Seattle, WA; Department of Pharmacy (JSM, DKB, SDS), University of Washington, Seattle, WA; Cornerstone Systems (LC), Lynden, WA; and Amgen, Inc (JLM), Thousand Oaks, CA. Dr Malin is now with the Veterans Affairs Greater Los Angeles Health Care System, Los Angeles, CA.

 Drs Ramsey and Sullivan report receiving a grant from Amgen, Inc for this research. Ms Clarke reports serving as a paid consultant to the Fred Hutchinson Cancer Research Center for her involvement in this study. Dr Malin was an employee of Amgen, Inc at the time this work was performed and now serves as a paid consultant to the company. The other authors (JSM, DKB, CLM) report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Concept and design (SDR, JSM, JLM, SDS); acquisition of data (SDR, CLM, LC); analysis and interpretation of data (SDR, JSM, DKB, LC, JLM, SDS); drafting of the manuscript (SDR, JSM, CLM, SDS); critical revision of the manuscript for important intellectual content (SDR, JSM, DKB, CLM, JLM, SDS); statistical analysis (SDR, DKB); provision of study materials or patients (SDR); obtaining funding (SDR, SDS); administrative, technical, or logistic support (SDR, JSM, CLM, LC); and supervision (SDR). Additional Information: References 31 through 54 are cited in eAppendix B.

Articles

We linked health insurance records to cancer registry data to analyze colony-stimulating factor use, finding wide divergence from that recommended by practice guidelines.

Colony-Stimulating Factor Prescribing Patterns in Patients Receiving Chemotherapy for Cancer

: To assess factors that may influence men’s preference for surgery versus nonsurgical options among newly diagnosed patients considering treatments for local-stage prostate cancer.

: Prostate cancer patients were approached at urology clinics after diagnosis but prior to starting treatment in California, South Carolina, and Texas. Using a survey about the treatment decision-making process, patients were asked about their likes and dislikes of 5 common treatment options: surgery (prostatectomy), brachytherapy, external beam radiation therapy, hormone therapy, and watchful waiting.

: Logistic regression identified associations between treatment characteristics and choice of prostatectomy compared with nonsurgical options, controlling for demographic, clinical, and psychological covariates.

: Of the 198 eligible men who returned the baseline survey, 59% indicated they only considered surgery and 41% considered at least 1 nonsurgical option. In multivariate analysis, patients who thought treatment efficacy was a primary concern were significantly more likely to prefer surgery only (odds ratio [OR] = 6.20, 95% confidence interval [95% CI] = 1.74, 22.10); those indicating concern about personal burden were significantly more likely to prefer nonsurgical options (OR = 0.07, 95% CI = 0.02, 0.22). Advice of friends and relatives and concerns over side effects were not significantly associated with preference for surgery versus other treatments.

: Men’s perceptions about treatment efficacy and the personal burden of treatment dominated preferences for surgery versus nonsurgical options. Interventions to aid treatment decision making should account for these elements to minimize the impact of physician biases and patient misperceptions on men’s decisions as how best to manage their prostate cancer.

We conducted a multisite survey of men with local-stage prostate cancer to understand their preferences for available treatments. The most influential factors were perceptions of efficacy and personal burden of prostatectomy relative to nonsurgical options.

Men misperceived the efficacy (cure) of prostatectomy relative to other options.

Personal burdens associated with treatment (eg, recovery time) were more influential than long-term side effects.

Policymakers and clinicians should consider wider use of decision aids, which have been shown to improve men’s understanding of management options, and may counter what appear to be biases influencing the decision processes.

Most men with prostate cancer are diagnosed at early stage and have several treatment options available, including prostatectomy, external beam radiation therapy, brachytherapy, and watchful waiting. Overall survival following any management choice approaches 99% at 5 years.

 A recent review found no compelling evidence favoring one approach versus another.

 Despite the lack of consensus about treatment, prostatectomy is commonly selected by men with early-stage prostate cancer, particularly those with low-risk disease.

 The reasons for this preference are unclear. To understand why men favor prostatectomy versus other options, we conducted a prospective survey of a diverse sample of men. Participants included men newly diagnosed with local-stage prostate cancer who were considering their treatment options.

We based our model on the taxonomy described by Holmboe and Concato, who categorized treatment attributes (likes and dislikes) into 4 axes: (1) external information, (2) intrinsic characteristics of treatment, (3) personal impressions, and (4) economic concerns.

 Using this taxonomy, we conducted focus groups with prostate cancer patients to develop relevant attributes for each management approach (including watchful waiting) that patients considered important. Based on our focus groups and literature review, we grouped characteristics into 5 domains: personal beliefs, treatment efficacy, others’ experience, personal burden, and avoiding side effects (see the 

We hypothesized that patients who indicated 

 as a main concern would be more likely to prefer prostatectomy only. Second, we hypothesized that men who indicated that a 

 influenced their decision (for reasons other than perceived efficacy) would favor surgery because perceptions about curing other common cancers involve removing the tumor (eg, breast cancer). Third, persons who listed 

 as highly important to their decision would be less likely to prefer prostatectomy, as surgery has been associated with higher rates of incontinence and impotence compared with other options.

experience of a family member or friend with prostate cancer 

who had surgery would be more likely to prefer prostatectomy versus nonsurgical options. Patients who have undergone surgery have been reported to be strong advocates for surgery,

 and many men choose surgical options, which increases the “exposure” to this management option. Finally, we hypothesized that patients concerned with the 

 of a treatment would prefer nonsurgical options because of the family and time burdens associated with recovery and managing a catheter. We also examined associations between treatment choice and general physical and emotional quality of life, prostate-specific symptoms, and the beliefs/perceptions patients had about prostate cancer and treatment.

Survey Development: Focus Groups and Cognitive Interviews

To develop questions that reflected the conceptual model, we conducted focus groups and cognitive interviews. Participants were identified through the Puget Sound Cancer Surveillance System, the Surveillance, Epidemiology and End Results (SEER) cancer registry for Western Washington State. Sampling randomly from SEER records for 2 urban counties, we contacted men diagnosed with local-stage prostate cancer within 6 months prior to the sampling date. Following physician notification to 69 patients, 14 men agreed to participate. Prior to the focus groups, participants received a set of likes and dislikes about treatment options based on the conceptual model.

 During the discussions, focus group participants ranked likes about the treatment they received and dislikes about treatments they did not select. Based on focus group feedback, we reduced the items in the final survey to 10 reasons (likes) for considering each treatment and 14 reasons (dislikes) for not considering a treatment. Survey items listing reasons for considering a treatment included doctor recommendation and opinions that the treatment would be more effective, have the fewest side effects, or have the least impact on the patient’s life. Survey items listing reasons for not considering a treatment included the converse of the previous 3 items and insurance not covering the treatment or out-of-pocket costs being too high. See the Appendix for the complete list.

We also included items from previous surveys developed for prostate cancer patients as well as surveys assessing general health,

 (see the Appendix). Using identical patient recruitment methods, these questions and the revised items were presented to 9 men in one-on-one cognitive interviews. The objectives were to confirm that patients understood the survey items and to assess survey fatigue. During the cognitive interviews, the participants completed semifinal versions of the survey while discussing the responses with an interviewer.

Participants were recruited through informed consent procedures approved by an institutional review board and were reimbursed $25 for their involvement.

Patients were recruited for the Family And Cancer Therapy Selection (FACTS) study. This study was initiated to understand the full range of domains involved in decision making by patients and partners, and how they make prostate cancer treatment decisions together. Newly diagnosed patients were approached in urology clinics at 3 sites: the Medical University of South Carolina (the University Urology Department and Ralph H. Johnson Veterans Administration Medical Center, Charleston), the University of Southern California (3 university-affiliated medical centers in Los Angeles), and the University of Texas Health Science Center at San Antonio (University Physicians Group and Audie Murphy Veterans Administration Hospital). Eligible patients included those diagnosed with local-stage disease (prostate-specific antigen [PSA] <50; tumor, node, metastasis [TNM] stages T1-T2, N0, M0) who had not initiated treatment other than hormone therapy at the time of approach. Clinic study coordinators reviewed lists of newly diagnosed local-stage prostate cancer patients scheduled for clinic visits. Interested patients signed consent forms and received a take-home survey to return by mail. Patients who indicated they had started treatment prior to the survey return date were excluded from this analysis (n = 42).

Study materials were approved by the institutional review board at each accrual site and the coordinating center. Participants received $25 after completing the baseline survey.

We conducted bivariate analyses of the relationships with each covariate and the primary outcome “preferring surgery only versus preferring at least one nonsurgical option.” We collapsed the outcome variable rather than keep individual nonprostatectomy categories because of the small number of observations in the nonprostatectomy categories. Chi-square tests were used for all categorical variables, and 

 tests were used to assess bivariate relationships for continuous variables. We used logistic regression to assess the association between our hypothesized likes and dislikes, and preferring only surgery versus nonsurgical options, adjusting for the demographic, clinical, and patient psychological factors noted in the Results section. We used a fixed-effects model, including recruitment site as the fixed effect in the model. This allowed for a withinsite assessment of the other variables. One limitation of this approach is the potential for other, unmeasured variables to confound the primary association between the outcome and our hypothesized likes and dislikes. Adjusting for the demographic, clinical, and patient psychological factors in the Results section addressed confounding to some extent. However, we recognize that there may be variables we were not aware of that could affect our results. The Hosmer-Lemeshow goodness-of-fit test was used to assess the overall adequacy of the model. All variables were retained in the model to assess their joint association on the outcome. For the continuous variables including the cancer worry scale,

 the Expanded Prostate Index Composite scores,

 nonparametric smoothing was used to assess the unadjusted functional form between each of these variables and choice of surgery. For all these variables, the functional form was approximately linear; thus, each of these variables was entered linearly into the logistic regression model. No adjustments were made for multiple tests. All tests for statistical significance used a 

A total of 423 patients were approached for participation. Of these, 240 met eligibility criteria and 198 (83%) returned the survey prior to initiating treatment; the analysis included the 198 patients (

). The median time from diagnosis to participant survey return was 70 days. The average age of the patient at the time of diagnosis was 63.0 years (SD = 8.1 years) and 36% were under the age of 60; 71% were white, 59% were employed, and 48% were college graduates (

). Eighty-six percent had a PSA level of <10, 62% had a Gleason score of <6 at diagnosis, and 55% were categorized as low risk based on the combination of these factors.

 Thirty-six percent listed 2 or more noncancer comorbidities at diagnosis (Table 1).

Overall, 116 (59%) patients indicated they were considering only surgery. More than three-quarters (76%) of all patients were only considering a single treatment option, with 12% considering 2 treatments, 8% considering 3 treatments, and 2% considering 4 or 5 treatment options. We examined all options considered by patients, including those considering multiple options: 71% reported they were considering prostatectomy, 19% brachytherapy, 15% external beam radiation therapy, 15% watchful waiting, and 8% hormone therapy (data not shown).

On bivariate analysis, men preferring prostatectomy only were more likely to be less than 70 years old, have private insurance, be married, and be employed compared with those considering at least 1 nonsurgical option (Table 1). Men preferring prostatectomy scored higher on the SF-12 physical component, but not the mental component, than patients preferring other options. Sexual function was significantly higher for men preferring prostatectomy only (Table 1).

The survey domains were commonly ranked as important to the treatment decision. Nearly half (45%) of the patients considering nonsurgical options reported personal burden as important to their decision compared with 10% of men only considering surgery (

). The experience of others was more likely to influence men considering surgery only. Both groups listed avoidance of side effects and personal beliefs nearly evenly in their likes and dislikes about treatment options; treatment efficacy was more commonly listed by surgeryonly men as important to their decision (see the Appendix).

Men preferring prostatectomy only were no different from those preferring other options regarding their trust in the healthcare system, levels of prostate cancer anxiety and worry, and feelings of cancer control and self-efficacy (Table 2). For those preferring prostatectomy only, there was no significant difference in the likelihood of receiving multiple different treatment recommendations compared with men considering other options.

Men concerned about the personal burden of treatment were significantly more likely to prefer nonsurgical options, and men who gave weight to the experiences of others were more likely to prefer prostatectomy only. Those concerned with treatment efficacy were significantly more likely to consider only prostatectomy. There were no significant differences in preferences for prostatectomy or nonsurgical options between those who indicated that personal beliefs or avoiding side effects was a key factor in their treatment decision.

Of the hypothesized likes and dislikes explored in the multivariate logistic regression analysis, treatment efficacy and personal burden remained strongly associated with treatment preference after adjusting for demographic, clinical, and psychological factors (

). The experience of others was not significantly associated with treatment preference. Those who indicated that treatment efficacy was a main reason for their treatment choice were significantly more likely to prefer surgery only (odds ratio [OR] = 6.20, 95% confidence interval [95% CI] = 1.74, 22.10). Those who considered personal burden a top reason for their treatment preference were much less likely to prefer surgery (OR = 0.07, 95% CI = 0.02, 0.22). General and prostate-specific measures of quality of life were not significantly associated with preferring surgery only. In the multivariate model, Hispanics and Asians were more likely to only consider surgery compared with whites (OR = 10.77, 95% CI = 1.77, 65.33).

lists the top reasons for considering and not considering each treatment choice. Patients considering brachytherapy most often believed it had the fewest side effects. The most common reason stated by those considering external radiation was research they had done on their own. The most common reason men stated for not considering prostatectomy, brachytherapy, external radiation, and/or hormone therapy was that the doctor recommended other treatment(s) (Table 4). The most common reason for not considering watchful waiting was fear about the cancer spreading.

To better understand the decision-making process for men who have been diagnosed with local-stage prostate cancer, we conducted a prospective, multicenter study of patients prior to initial therapy. We asked participants to report the management options they were considering and their reasons for liking and disliking each option. Our primary aim was to identify differences in likes and dislikes that influence treatment choice, focusing on differences between men preferring prostatectomy only and men preferring at least 1 nonsurgical option. Approximately 59% of men preferred only prostatectomy. We identified 2 strong influences on patient preferences. First, men who cited a strong concern for treatment efficacy were significantly more likely to prefer surgery. Second, men concerned with personal burdens of treatment were significantly more likely to prefer nonsurgical options. Notably, factors including disease risk among this population with local-stage cancer, side effects, and quality-of-life considerations did not emerge as strong predictors of preference for or against only considering surgery.

Prior studies of prostate cancer decision making surveyed men weeks to months after they had started or completed therapy.

 The FACTS study is unique because men provided insight into their treatment preferences as they were making their decision. A second feature of our study is our attempt to include a racially diverse cohort. Interestingly, we found that Hispanics and Asians, both of whom have a lower likelihood of prostate cancer than whites and blacks, are much more likely to prefer surgery only. This finding is consistent with a study of Hispanic men by Denberg and colleagues, who found that Latinos received prostatectomy more often than whites.

 We are not aware of studies of decision making among Asian men.

Many men in our sample appeared to believe that prostatectomy has superior efficacy compared with other options. This finding has been observed in multiple studies among men who were asked to recall reasons for their treatment choice,

 but is not consistent with current evidence.

 In contrast to our hypothesis about side effects and decisions, and despite evidence showing differential risk for side effects among management options,

 we found that concern about side effects appeared not to influence men’s preferences for surgery and consideration of nonsurgical options. This finding, combined with the apparent positive relationship between men’s concern about efficacy and preference for prostatectomy, suggests that many men have either limited access to or a poor understanding of the evidence regarding management options for local-stage prostate cancer. To address this issue, medical professionals who support prostate cancer patients during their decision process may need to take additional measures to ensure patients have a clear and balanced picture of the risks and benefits of all management options. Decision aids (eg, online or in the form of DVDs) have been shown to help patients make more informed decisions about treatment.

Many men in this study expressed concern with the impact of treatment on daily life (ie, personal burden). Perceived treatment burdens had a substantial negative impact on patients’ willingness to undergo prostatectomy. Holmboe and Concato also found burden of treatment to be a powerful factor in treatment decision making.

 Gwede and colleagues noted that persons who chose options other than prostatectomy cited burden issues as reasons to avoid surgery.

 These findings may explain the recent popularity of robotic prostatectomy, which has been touted to reduce recovery time compared with open prostatectomy.

The results of the multivariate analysis did not support our hypotheses regarding the influence of affected friends or family members on patients’ decisions. In contrast to our study, Diefenbach and colleagues found that the influence of fellow patients was second only to physicians in the decision process.

 It is possible that our sample included men who were different in some respects from men in prior studies and who were less likely to consider the opinions of friends and family. This possible difference may be due to the academic environment of our study sites, where second opinions are common.

We note some limitations of the FACTS study. We conducted it in academic urology clinics in 3 states. Although there was considerable diversity in this sample, our findings may not be generalizable to other settings. Additionally, many of the men in the sample (56%) were seeking second opinions. Our sample, therefore, may include a cohort of self-selected men with a preference for surgery. Although this factor was not an important predictor of surgery preference in the multivariate model, we acknowledge that our findings may be unique to urology settings with many patients seeking a second opinion. Second, although all patients were approached prior to treatment and the majority of men in our study returned their surveys prior to starting therapy, a portion returned surveys after initiating treatment. Although we excluded from the analysis men who had started treatment, it is possible that men who received treatment soon after diagnosis may have different perceptions than men who completed the survey prior to treatment. Third, a substantial proportion of our participants favored prostatectomy as their sole treatment consideration, thus limiting our statistical power to detect differences among the various nonsurgical treatments. We did not differentiate between robotic or open surgical procedures. Future studies of patient decision making could oversample patients considering nonsurgical options to better delineate differences between those management strategies. Although we observed differences in beliefs about treatment efficacy and personal burdens, we did not specifically measure patient knowledge about prostate cancer treatment. Future studies should examine whether inaccurate perceptions about treatment persist among men who receive accurate information, as measured by consistency in knowledge about treatment outcomes.

Patients’ personal beliefs about the value of one treatment versus another, as well as their assessments of the efficacy, burden, and side effects of various treatments, are influenced by several factors. These factors include their doctor’s advice and information available through a variety of other sources such as pamphlets, books, and the Internet. Further research is needed to characterize what leads to men’s likes and dislikes of various treatment choices. In particular, the content of the physician—patient interaction and its influence on patient considerations need further evaluation.

The authors gratefully acknowledge the contributions made by Catherine Fedorenko, MMSc, and Megan Fairweather, BS, in preparing this manuscript.

From the Health Services Research & Development Center of Excellence (SBZ), VA Puget Sound Health Care System, Seattle, WA; Public Health Sciences Division (SBZ, CMM, SDR), Fred Hutchinson Cancer Research Center, Seattle, WA; School of Pharmacy (DKB), University of Washington, Seattle, WA; Department of Urology (DFP), University of Southern California, Los Angeles, CA; Division of Cancer Prevention and Control (IJH, JLS, DUE), Centers for Disease Control and Prevention, Atlanta, GA; Department of Urology (IMT), University of Texas Health Science Center, San Antonio, TX; and Division of Urology Services (TEK), Medical University of South Carolina, Charleston, SC. Dr Penson is now with the Vanderbilt Institute for Medicine and Public Health, Vanderbilt University Medical Center, Nashville, TN.

This publication was supported by Cooperative Agreement 1-U48-DP-000050 from the Centers for Disease Control and Prevention, Prevention Research Centers Program, through the University of Washington Health Promotion Research Center. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.

Case ascertainment for the focus group research was supported by the Cancer Surveillance System of the Fred Hutchinson Cancer Research Center, which is funded by Contract N01-PC-35142 from the Surveillance, Epidemiology and End Results (SEER) Program of the National Cancer Institute with additional support from the Fred Hutchinson Cancer Research Center and the State of Washington.

The authors (SBZ, CMM, DKB, DFP, IJH, JLS, DUE, IMT, TEK, SDR) report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

In a survey of local-stage prostate cancer patients, preference for prostatectomy was influenced by perceptions of its efficacy and personal burden versus nonsurgical options.


Preliminary Treatment Considerations Among Men With Newly Diagnosed Prostate Cancer

To examine the impact of a medication copayment increase on adherence to diabetes, hypertension, and hyperlipidemic medications.

Retrospective pre—post observational study.

This study compared medication adherence at 4 Veterans Affairs medical centers between veterans who were exempt from copayments and propensity-matched veterans who were not exempt. The diabetes sample included 1069 exempt veterans and 1069 nonexempt veterans, the hypertension sample included 3545 exempt veterans and 3545 nonexempt veterans, and the sample of veterans taking statins included 2029 exempt veterans and 2029 nonexempt veterans. The main outcome measure was medication adherence 12 months before and 23 months after the copayment increase. Adherence differences were assessed in a difference-in-difference approach by using generalized estimating equations that controlled for time, copayment exemption, an interaction between time and copayment exemption, and patient demographics, site, and other factors.

Adherence to all medications increased in the short term for all veterans, but then declined in the longer term (February-December 2003). The change in adherence between the preperiod and the postperiod was significantly different for exempt and nonexempt veterans in all 3 cohorts, and nonadherence increased over time for veterans required to pay copayments. The impact of the copayment increase was particularly adverse for veterans with diabetes who were required to pay copayments.

A $5 copayment increase (from $2 to $7) adversely impacted medication adherence for veterans subject to copayments taking oral hypoglycemic agents, antihypertensive medications, or statins.

Adherence to diabetes, hypertension, and hyperlipidemic medications among veterans exempt from copayments and propensity-matched nonexempt veterans was examined at 4 Veterans Affairs medical centers.

A $5 copayment increase (from $2 to $7) adversely impacted medication adherence for veterans subject to copayments.

Copayment increases need to be considered carefully by the Department of Veterans Affairs to ensure that veterans who have greater comorbidity and lower incomes than the general US population do not forgo needed medications.

Diabetes and hypertension are among the most common chronic conditions in the United States, with prevalence rates of 9.6% and 29.7% for individuals age 20 years and over in 1999-2001.

 Between 20% and 40% of patients with these conditions receive no medication management.

 Pharmacologic therapy is a mainstay in the management of both conditions, but only 50% to 70% of patients receiving medication management are adherent to medications.

Improving medication adherence for individuals with diabetes or hypertension has been challenging as health plans and employers increase medication copayments,

 lower limits on the number or total reimbursement of covered medications,1

Between 2000 and 2005, average copayments for commercially insured individuals increased from $7 to $10 for generic medication, $13 to $22 for preferred medications, and $17 to $35 for nonpreferred medications.

 Cost-related nonadherence has increased with higher copayments.

The Department of Veterans Affairs (VA) copayment policy mirrored these market trends by increasing medication copayments from $2 to $7 for a 30-day fill on February 4, 2002. In 1999-2000, 19.6% of veterans had diagnosed diabetes

 Lipid-lowering medications are indicated for nearly all of these veterans. This study examined the impact of the VA medication copayment increase on adherence to diabetes, hypertension, and hyperlipidemic medications by veterans with diabetes or hypertension during a 35-month period (February 2001-December 2003).

This analysis contributes to the extensive literature on the effect of copayments on medication adherence by assessing several therapeutic classes across several conditions, allowing us to examine whether copayment increases have a differential impact across conditions. In addition, we assessed a copayment increase in a population with higher comorbidity and lower income than those in many prior studies, which is critical to clarify whether copayment impacts vary by income.

A recent review article of interventions to impact medication adherence found that many evaluations of formulary and cost-sharing changes lacked control groups or pre—post comparisons, which limited their internal validity.

 We included a colocated control group that controlled for site effects not included in other studies,

 compared adherence before and after the copayment increase,

 and reduced potential bias from the nonequivalent control group with propensity score matching and covariate adjustment that could have confounded copayment effects in prior studies.

 This longitudinal comparison provides information on additional decrements in adherence that might occur with further copayment increases in the VA or private insurers. Identifying patients with chronic conditions who might be especially adversely impacted by copayment increases also could suggest targets for interventions to offset the adherence impacts of increased copayments.

The systemwide increase in the VA medication copayment

 from $2 to $7 in February 2002 created a natural experiment to examine changes in medication adherence for veterans with diabetes or hypertension. Prior to this medication copayment increase, the VA implemented $15  copayments for primary care, $50 copayments for specialty care, and $10 per diem copayments for inpatient care effective December 6, 2001.

 On January 1, 2006, the medication copayment was increased again to $8 for a 30-day fill.

We used a retrospective, pre—post cohort design with a nonequivalent, colocated control group at 4 large tertiary Veterans Affairs medical centers (VAMCs). We identified 60,017 veterans with diabetes (n = 23,182) or hypertension (n = 51,503) who were diagnosed and prescribed a medication for either of these conditions in 2000. Veterans were included in the analysis if they (1) were alive during the entire study period, (2) had a majority of their primary care visits at 1 of the 4 VAMCs, (3) had complete information on level of military service–connected disability to determine exemption from drug copayments, (4) were not hospitalized when the copayment increase went into effect or for more than 1 year during the study period, (5) had at least 1 fill in a relevant drug class during the quarter prior to the copayment change, and (6) had at least 1 fill during the second, third, and fourth quarters prior to the copayment change.19-23 We excluded subjects on non-NPH insulin therapy that would preclude taking oral hypoglycemic agents. We did not exclude patients on NPH, because NPH insulin may be added to oral regimens in a stepped approach. The application of these criteria resulted in analytic samples of 7852 veterans with hypertension, 4407 veterans with diabetes, and 4217 veterans with diabetes or hypertension who were taking statins.

A veteran’s obligation to pay medication (and healthcare) copayments is determined by priority group assignment, based on military service-connected disability for each diagnosed condition, and on income. In 2002, veterans were exempt from medication copayments if (1) their annual income was less than $9556 if single and $12,516 if married; (2) their diabetes, hypertension, or hyperlipidemia was a service-connected disability; or (3) their diabetes, hypertension, or hyperlipidemia was not a service-connected disability, but they exceeded the $840 copayment cap in a given year. Priority group 1 veterans are exempt from all healthcare and medication copayments because the VA has determined that 50% or more of their overall disability is due to their military service, whereas priority group 7 and 8 veterans are required to pay all healthcare and medication copayments because they have no military service—related disability and have income and/or net worth above the VA national income threshold.We excluded veterans in priority groups 2 to 6 from the study because we were unable to determine whether they were required to pay medication copayments.

Unadjusted comparisons of exempt and nonexempt veterans demonstrated significant differences in every observed characteristic. To reduce potential bias from imbalance in observed covariates between exempt and nonexempt veterans and to improve equivalence of the control groups, we conducted 1-to-1 nearest-neighbor propensity score matching with replacement.

 After running 3 logistic regressions to generate propensity scores and matching exempt and nonexempt veterans, 762 exempt veterans with hypertension, 317 exempt veterans with diabetes, and 159 exempt veterans taking statins were excluded because there were no nonexempt veterans with similar propensity scores.

Our final hypertension matched sample included 3545 exempt veterans and 3545 nonexempt veterans. Our final diabetes matched sample included 1069 exempt veterans and 1069 nonexempt veterans. Our final matched sample of veterans with diabetes or hyperlipidemia taking statins included 2029 exempt veterans and 2029 nonexempt veterans. The unit of analysis was person-month with each veteran having up to 35 repeated measures. Human Subjects committees for all coinvestigators’ facilities (Ann Arbor, MI, Durham, NC, Hines, IL, Little Rock, AR, and Seattle, WA, VAMCs) reviewed and approved this study.

We used 4 VA datasets for 2001-2003. All medications dispensed from the VA are recorded in the national Pharmacy Benefits Management database; data elements include drug name, date dispensed, number of days of medication supplied, and dosage.

 The VA inpatient and outpatient care files provided information on veteran demographic characteristics and diagnoses for every inpatient hospitalization and outpatient visit in the national VA system. Benefit Identification and Record Locator System death record data identified which veterans died during the study period. Finally, the Diagnostic Cost Group Hierarchical Cost Category (DCG/HCC) version 6.0 score was used to adjust for overall comorbidity; this measure has been shown to predict veterans’ total costs

Prescription Drug Use and Assessment of Medication Adherence

We calculated monthly medication adherence using the validated ReComp algorithm, a modification of a widely used method that is correlated with a variety of clinical outcomes.

 This algorithm estimates the proportion of days covered for a given measurement interval using the date dispensed and the number of days supplied with each fill. Subjects were considered adherent if they had medications available for at least 80% of each month, which is a conventional threshold that was used to maintain congruence with prior studies.

For adherence to oral hypoglycemic agents (OHAs) (sulfonylureas, metformin, thiazolidinediones) and antihypertensive medications (angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, calcium channel blockers, beta-blockers, and alpha-1 antagonists), we calculated refill adherence separately for each class of medications. We averaged scores to produce a monthly composite OHA adherence score among the diabetes cohort and a monthly composite antihypertensive adherence score among the hypertension cohort. Adherence in veterans with diabetes and/or hypertension taking HMG-CoA reductase inhibitors (statins) was calculated by combining all statin drugs in a single adherence measurement.

There were 3 explanatory variables of interest in the medication adherence analysis using the propensity-matched cohorts: (1) an indicator of whether a veteran was required to make copayments; (2) time indicators for the 12-month preperiod before the copayment increase (February 2001-January 2002), the 12 months (short-term postperiod) just after the copayment increase (February 2002-January 2003), and the subsequent 11-month longer-term postperiod (February 2003-December 2003); and (3) an interaction of the copayment exemption and time indicators to enable a difference-in-difference analysis. The postperiod was subdivided to examine whether adherence differed in the short term and longer term.

All models also were adjusted for age, sex, race, marital status, patient comorbidity measured as DCG/HCC score, whether a veteran was hospitalized in prior or current months, presence of a depression diagnosis at baseline, presence of comorbid diabetes (if hypertension cohort) or hypertension (if diabetes cohort), and the number of all other medications that the patient was prescribed during the preperiod. To adjust for the impact of the December 2001 outpatient visit copayment increases, we adjusted for the number of primary care, specialty care, and mental health visits 90 to 180 days prior to the current month. It is possible that the increase in healthcare copayments would decrease outpatient visits and decrease prescription renewals; the lagged visit counts attempted to control for these cross-price effects.50 Models that included a pre—post indicator for initiation of the healthcare copayment generated similar results (results not presented).

Logistic regressions were estimated to identify whether a veteran was exempt or nonexempt from copayments and to generate predicted probabilities of being nonexempt, which served as the propensity scores for matching exempt and nonexempt veterans in the diabetes, hypertension, and statin cohorts. The propensity score model for the diabetes cohort included 19 main effects related to demographics, comorbidity, and medication burden and 18 interactions, because this specification reduced covariate imbalance within propensity score quintiles better than other specifications. Similar iterative specification tests were conducted for the logistic regressions on the hypertension cohort with a final specification including 11 main effects and 18 interactions, and the statin cohort with a final specification including 20 main effects and 28 interaction terms. Detailed results are given in 

Bivariate statistics (t tests, χ2 test) were estimated to compare patient characteristics between matched exempt and nonexempt veterans, and to compare medication adherence at baseline (February 2001) and the last month of the study period (December 2003). To examine pre—post differences between propensity-matched exempt and nonexempt veterans in medication refill adherence over 35 months (12 preperiod and 23 postperiod months), we used generalized estimating equations assuming a binomial distribution, logit link, and independent working covariance structure with person-month as the unit of analysis. The working covariance structure was specified to obtain unbiased parameter estimates for the timevarying covariates.

To estimate first differences, predictions from these generalized estimating equations were obtained to identify the proportion of exempt and nonexempt veterans in each cohort who were adherent in the preperiod, the immediate postperiod, and the longer-term postperiod. Confidence intervals (CIs) for these proportions were estimated through 1000 bootstrap iterations. All analyses used Stata software (StataCorp, College Station, TX). All reported P values are 2-sided, and significance was lowered to P = .01 to account for multiple comparisons.

Propensity matching eliminated imbalance in several covariates (proportion of veterans with diabetes, hypertension, or both; marital status, sex, race, baseline hospitalization rates, and number of diabetes medications) across all 3 cohorts, but did not eliminate imbalance in site composition (

). Nonexempt veterans with diabetes were slightly older than exempt veterans with diabetes (

 <.001) and were taking more hypertension medications (

 <.001) but fewer medications of all kinds (

 <.0001). Nonexempt veterans with diabetes had slightly fewer primary care and specialty care visits on average than exempt veterans (both 

Nonexempt veterans with hypertension were older than exempt veterans with hypertension (

 <.0001), had slightly higher DCG/HCC risk scores (

 <.01), and fewer specialty care visits (

 <.0001). Nonexempt veterans with diabetes and/or hypertension taking statins were older than exempt veterans taking statins (

 <.0001), were taking fewer medications overall (

 <.001), had slightly fewer primary care, specialty care, and mental health visits on average (all 

 <.0001), and slightly fewer mental health visits (

The unadjusted proportion of nonexempt and exempt veterans with diabetes who were adherent to their OHAs at baseline (February 2001) was similar, but the proportion adherent in the last month of the study period (December 2003) was significantly lower among nonexempt veterans with diabetes (60% vs 69%; 

 <.0001). A greater proportion of nonexempt veterans with hypertension were adherent to their antihypertensive medication at baseline (80% vs 76%; 

 <.0001), but there were no differences in adherence in December 2003. Unadjusted adherence rates were similar between nonexempt and exempt veterans taking statins.

After covariate adjustment, OHA adherence among exempt veterans with diabetes increased 4.1 percentage points in the year after the copayment increase (February 2002-January 2003) compared with the preperiod, while adherence remained constant for nonexempt veterans subject to copayments (first difference −3.8%; 95% CI = −3.7%, −3.9%). In the longer-term postperiod (February-December 2003), OHA adherence declined for both diabetes cohorts compared with the preperiod but significantly more so for nonexempt veterans (−10.3% vs −0.9%; 

 <.001) (first difference −9.6%; 95% CI = −9.5%, −9.8%). See 

Adherence to antihypertensive medications increased for exempt and nonexempt veterans in the year after the copayment increase (4.1% vs 5.9%; first difference −1.8%; 95% CI = −1.8%, −1.9%) compared with the preperiod, but decreased thereafter for both groups compared with the preperiod. The decline in adherence to antihypertensive medications was greater for nonexempt veterans (−5.4% vs −2.3%; first difference −3.2%; 95% CI = −3.1%, −3.3%). See 

Adherence to statins increased for exempt and nonexempt veterans in the year after the copayment increase compared with the preperiod, but more so for exempt veterans (3.5% vs 6.6%; first difference −3.0%; 95% CI = −2.9%, −3.1%). Statin adherence continued to increase (1.2%) for exempt veterans in the longer-term postperiod (February-December 2003) compared with the preperiod, but decreased for nonexempt veterans (−1.9%; first difference −3.1%; 95% CI = −3.0%, −3.2%). See 

Adherence to OHAs, antihypertensive medications, and statins by veterans with diabetes or hypertension increased in the year after a $5 medication copayment increase (from $2 to $7), but subsequently declined. The change in adherence between the preperiod and postperiod (as indicated by the first differences) was significantly different for exempt and nonexempt veterans in all 3 cohorts, which indicates that the medication copayment increase had adverse effects on medication adherence, as has been found in similar studies. The longer-term impact of this policy change was particularly adverse for veterans with diabetes who were required to make copayments, because their adherence to OHAs 13 to 23 months after the copayment increase was 10.3% lower than their preperiod adherence and 9% lower than the adherence of comparable veterans who were exempt.

The initial increase in adherence may have been due to VA physicians encouraging all of their patients (regardless of copay status) to stockpile medications in anticipation of the copayment increase, because VA physicians tend not to be aware of a veteran’s copayment exemption status and to treat all veterans in their panel similarly. This initial increase and subsequent decline also could have been due to a lagged effect of the copayment increase on adherence, because veterans may initially refill medications until their budgets get stretched and cost-related nonadherence ensues.

Cost-related nonadherence may have been reinforced by the cross-price effect

 of the increased primary care copayment (from $0 to $15) and specialty care copayment (from $15 to $50) that occurred 2 months prior to the medication copayment increase. Increased healthcare copayments could impact adherence rates in 2 ways. The out-of-pocket costs of outpatient visits could have reduced visit rates at which prescriptions would be initiated, modified, or renewed. (In the postperiod, we observed significantly lower rates of primary care visits for veterans in all 3 cohorts [

 <.0001].) In addition, veterans with fixed incomes who were keeping their outpatient appointments might sacrifice their medications to pay for these visits.

 It also is possible that the initial adherence increase and subsequent decline could have been an artifact of our inclusion criteria. In a sensitivity analysis, we included veterans with 2 or more fills in the 12 months prior to the copayment increase (instead of 1 fill in the 3 months prior and another fill 4-9 months prior), but the difference-in-difference results were unchanged because trends were similar for exempt and nonexempt veterans.

The decline in adherence to OHAs was significantly greater for veterans required to pay copayments than for exempt veterans, but the change in adherence to antihypertensives and statins was more modest. It may simply be that patients value their cardiovascular medications more than their diabetes medications, as has been shown in prior studies.

 A study of commercially insured populations by Goldman et al

A copayment increase from $2 to $7 adversely affected veterans' adherence to statins, antihypertensives, and oral hypoglycemic agents.


David K. Blough, PhD

Articles

Colony-Stimulating Factor Prescribing Patterns in Patients Receiving Chemotherapy for Cancer

Preliminary Treatment Considerations Among Men With Newly Diagnosed Prostate Cancer

Increasing Copayments and Adherence to Diabetes, Hypertension, and Hyperlipidemic Medications

Second-line and Third-line Chemotherapy for Lung Cancer: Use and Cost