United States Health Plan Research Articles

Analgesic opioid use in a health-insured epilepsy population during 2012.

Analgesic opioid use has increased dramatically in the general population. Although opioid analgesics are not indicated for the treatment of epilepsy, frequent opioid use has been reported in the epilepsy population. It is not clear whether comorbid disorders and/or epilepsy-associated injuries due to seizures foster opioid use. Our primary objective was to compare the prevalence of analgesic opioid use in an insured patient population with epilepsy to a matched control population without epilepsy. After observing increased frequency of opioid use in people with epilepsy compared with matched controls, we assessed the contribution of age, gender, pain diagnosis, and psychiatric illness as possible drivers regarding the use of opioids. Health insurance claims and membership data from nine United States (U.S.) health plans for the year 2012 were analyzed. Individuals with epilepsy (n=10,271) were match-paired at a 1:2 ratio to individuals without epilepsy (n=20,542) within each health plan using propensity scores derived from age group, gender, and insurance type. Matched comparison groups had 53% females and 47% males with an average age of 34 years for the group with epilepsy and 33 years for controls. Each matched comparison group included 66% of individuals with commercial insurance, 30% with Medicaid insurance, and 4% with Medicare coverage. Based on prescriptions filled at least once during 2012, prevalence of analgesic opioid use was determined. The percentages of individuals with diagnosis for specific pain conditions and those with psychiatric diagnoses were also determined for the two comparison groups. Analgesic opioids were used by 26% of individuals in the group with epilepsy vs. 18% of matched controls (p<0.001). Compared with matched controls, the group with epilepsy had a significantly higher percentage of individuals with all 16 pain conditions examined: joint pain or stiffness (16% vs. 11%), abdominal pain (14% vs. 9%), headache (14% vs. 5%), pain in limb (12% vs. 7%), chest pain (11% vs. 6%), sprain of different parts (9% vs. 7%), sinusitis (9% vs. 7%), migraine (8% vs. 2%), lumbago (8% vs. 6%), backache (6% vs. 4%), cervicalgia (6% vs. 3%), fracture (5% vs. 3%), fibromyalgia (4% vs. 3%), chronic pain (3% vs. 1%), sciatica (1.4% vs. 1%), and jaw pain (0.4% vs. 0.1%) (all p<0.001). The prevalence of pain diagnosis was 51% in the group with epilepsy and 39% in the matched control group (p<0.0001). The prevalence of 'psychiatric diagnoses' was 27% in the group with epilepsy and 12% in the matched control group (p<0.0001). The prevalences of analgesic opioid use, psychiatric diagnoses, and 16 pain conditions were significantly higher in the patient population with epilepsy than in the control population without epilepsy. Our study also showed how opioid use rate varied by gender, age category, and depression. The reasons for the greater prevalence of opioid use in people with epilepsy are unclear. It seems that increased pain prevalence is an important driver for the higher frequency of opioid use in people with epilepsy. Psychiatric illness and other factors also appear to contribute. Further analysis including more detailed clinical information that cannot be obtained through claims data alone will be required to provide more insight into opioid use in people with epilepsy. If opioid use is higher in people with epilepsy as our results suggest, physicians managing patients with epilepsy need to pay special attention to safe opioid prescribing habits in order to prevent adverse outcomes such as abuse, addiction, diversion, misuse, and overdose.

Incidence and prevalence of acromegaly in a large US health plan database

PurposeIncidence and prevalence estimates of acromegaly in the United States (US) are limited. Most existing reports are based on European data sources. The objective of this study was to estimate the annual incidence and prevalence of acromegaly in a large US managed care population, overall and stratified by age, sex, and geographic region, using data from 2008 to 2012.MethodsUsing administrative claims data, commercial health plan enrollees were identified with acromegaly if they had two or more medical claims with an acromegaly diagnosis code (ICD-9-CM: 253.0×) or one medical claim with an acromegaly diagnosis code in combination with one other claim for a pituitary tumor or pituitary procedure. The first date for an acromegaly-related claim set the index year. Incidence rates for each year were calculated by dividing the number of new acromegaly cases by the calculated person-time at risk. Annual prevalence estimates were calculated by dividing the number with any evidence of acromegaly by the total number of health plan enrollees enrolled for at least 1 day during each calendar year. Incidence and prevalence estimates were stratified by age (0–17, 18–44, 45–64, 65+ years), sex (male, female), and US geographic region of the health plan (Midwest, Northeast, South, West).ResultsOverall annual incidence rates of acromegaly were relatively constant across 2008–2012 with ~11 cases per million person-years (PMPY). Rates increased with age, ranging from 3–8 cases PMPY among children aged 0–17 years old to 9–18 cases PMPY among adults aged 65 and older. Females had 12 cases PMPY on average compared to 10 cases PMPY among men. On average, the Midwest had the lowest incidence rates (7 cases PMPY) compared to the Northeast, South and West (14, 12, and 10 cases PMPY, respectively). The overall annual prevalence of acromegaly was relatively constant across the 5 years from 2008 to 2012 with approximately 78 cases per million each year. Annual prevalence estimates increased with age, ranging from 29–37 cases per million among children aged 0–17 years old to 148–182 cases per million among adults aged 65 years and older. Males and females were similarly affected; each with approximately 77 cases per million each year. The Northeast and South had the highest prevalence estimates (92 and 89 cases per million, respectively); while the estimates for the West and Midwest were lower (65 and 57 cases per million, respectively) each year.ConclusionThis study examined 5 years of recent data to estimate the incidence and prevalence of acromegaly in a large geographically-diverse managed care population. The incidence rates were higher on average than published rates outside the US (11 vs. 3.3 PMPY), but prevalence estimates were consistent with previous reports. Incidence and prevalence both increased by age, did not differ for males and females, and varied slightly by US geographic region. The age and sex distribution of the selected population matched the known epidemiology of the disease. Using a claims-based approach, this analysis only captured acromegaly cases with an acromegaly-related medical claim. As a result, these estimates may underestimate the incidence and prevalence of acromegaly in US commercial health plans as they did not include individuals who were undiagnosed, in remission, undertreated, or not monitored during the study period. At the same time, these estimates may be viewed as an upper bound on the incidence of acromegaly in the US as the estimates did not include individuals who were in other health plans or uninsured during the study period. Additional evaluations are needed to identify the full extent of acromegaly in the US.Electronic supplementary materialThe online version of this article (doi:10.1007/s11102-015-0701-2) contains supplementary material, which is available to authorized users.

Nontuberculous Mycobacteria Lung Infections: Pre-index Comorbidity and Utilization Patterns at a Large United States Health Plan

Nontuberculous Mycobacteria Lung Infections: Pre-index Comorbidity and Utilization Patterns at a Large United States Health Plan

Use of Post-acute Care Services and Readmissions After Left Ventricular Assist Device Implantation in Privately Insured Patients

BackgroundVery little is known about health care resource utilization, including post–acute care use and hospital readmissions, after left ventricular assist device (LVAD) implantation. Methods and ResultsAdministrative claims from a database of multiple United States health plans were used to identify patients that received an LVAD (ICD-9 code 37.66) and survived to hospital discharge from January 1–2006, through September 30–2013. Post–acute care use was defined as a skilled nursing facility or rehabilitation stay within 90 days after hospital discharge. Patients were censored at heart transplantation or end of coverage through December 31–2013. Of 583 patients (mean age 55 years, 77% male), 223 (38.3%) used post–acute care services, more commonly in patients with diabetes, who required hemodialysis, and who had LVADs implanted at hospitals in more populated areas, with more beds, and in the northeast region (P < .05 for each). The most common reasons for readmission were device complications, heart failure, and arrhythmia. Readmission risk was higher in patients who had diabetes, peripheral vascular disease, and longer hospital length of stay, but it did not differ by post-acute care use. ConclusionsUse of post-acute care services varies based on hospital characteristics. We found no association between post-acute care use and readmission risk after LVAD implantation.

Background and rationale for the Society of Behavioral Medicine's position statement: expand United States health plan coverage for diabetes self-management education and support.

The Society of Behavioral Medicine (SBM) recognizes that diabetes self-management (DSM) education and support are fundamental to teaching people how to manage their diabetes and decrease disease-related complications. Implementation of the Patient Protection and Affordable Care Act provides an opportunity to expand DSM education and support to many people who are currently excluded from such services due to lack of insurance coverage, current policy barriers, or simple failure of healthcare systems to provide them. Extending the range and provision of such services could translate into reduced diabetic complications, a reduction in unnecessary healthcare utilization, and significant health-related cost savings on a national level. SBM recommends that public and private insurers be required to reimburse for 12h of DSM education and support annually for anyone with diabetes. Further, SBM recognizes that a range of modes and providers of DSM education and support have been shown effective, and that patient preferences and resources may influence choice. To address this, SBM urges health organizations to increase and diversify approaches toward DSM education and support they offer.

Symptom report and treatment experience of hypogonadal men with and without type 2 diabetes in a United States health plan.

Although hypogonadism (HG) is interrelated with type 2 diabetes mellitus (T2DM), there is little information about men's experiences with HG, T2DM, and testosterone replacement therapy (TRT). We examined symptoms and TRT use among men with HG, with and without T2DM, who received care within a single United States health plan. Men aged ≥ 18 years with HG, with and without T2DM, were identified from the 2008 to 2010 Reliant Medical Group electronic medical record database. Surveys responses compared by T2DM status using chi-square or Wilcoxon rank sum tests. A total of 93 men were included (19 with HG and T2DM, 74 with HG only). Men with both HG and T2DM were more frequently treated for their HG by an endocrinologist (52.6%), compared with men with HG only (28.4%, p = 0.058). Erectile dysfunction (ED) was the primary reason for seeking care among all surveyed men, although men with HG and T2DM reported experiencing ED more often (94.7%) than men with HG only (46.0%, p < 0.0001). Additional reasons for seeking care were similar between cohorts and included loss of energy and decreased sex drive. Most men (88.2%) reported using TRT, primarily as injection or gel formulations. Discontinuation of TRT was reported slightly more frequently by men with HG and T2DM (68.4%), compared with men with HG only (55.4%, p > 0.05). This study provides information about symptoms and TRT utilisation among HG men with and without T2DM. Men with HG only were less likely than those with both HG and T2DM to report that they were currently experiencing key symptoms compared with when they were first diagnosed, regardless of TRT utilisation, and were less likely to report ED as a current symptom.

Budgetary impact of adding ziv-aflibercept to a United States health plan formulary as a post-oxaliplatin biologic option for patients with metastatic colorectal cancer (mCRC).

765 Background: To assess the economic impact of the availability of ziv-aflibercept for patients with mCRC following disease progression after oxaliplatin, a user-adaptable budget impact model was developed for payers, based on use of 5-fluorouracil, leucovorin and irinotecan (FOLFIRI) chemotherapy alone or combined with a biologic agent, including ziv-aflibercept. Methods: A Markov-like model simulated transition from progression-free survival (PFS) to progressive disease (PD) and death. Over the model time frame of 1 year, each 90-day cycle adjusted for the probability of patients experiencing PFS, PD or death. Treatment options were FOLFIRI alone or combined with a biologic agent, ziv-aflibercept, bevacizumab, cetuximab or panitumumab. April 2013 United States utilization rates were used as baseline, with biologic agent drug costs as per July 2014 Medicare Part B Drug Average Sales Price. PFS and overall survival were calculated from clinical trial data of biologic agents combined with chemotherapy. Adverse event cost calculations used documented incidence rates and published cost sources. Other costs were for biomarker testing, drug administration/monitoring and death/terminal care. Results: The estimated number of patients receiving FOLFIRI alone or with a biologic agent for mCRC was 57 per million-member health plan. Using April 2013 utilization rates (including ziv-aflibercept at 3.4%), estimated treatment costs were $5.259M, a per-member-per-month (PMPM) cost of $0.0438. Changes in utilization rates of biologic agents, including, for example, ziv-aflibercept at 20.1%, would decrease costs by 2.1% ($108,176) to $5.151M annually, a PMPM decrease of $0.009. The model suggests cost savings primarily due to lower unit costs for ziv-aflibercept and lower death-related costs, rather than decreased adverse event management. Conclusions: The model estimates a PMPM cost saving with increased utilization of biologics, including ziv-aflibercept. Sensitivity analysis of utilization rate changes supports these findings. Including further user-defined scenarios allows plan-specific modeling of budgetary impact.

Background and rationale for the Society of Behavioral Medicine’s position statement: expand United States health plan coverage for diabetes self-management education and support

Background and rationale for the Society of Behavioral Medicine’s position statement: expand United States health plan coverage for diabetes self-management education and support

Trends in elective labor induction for six United States health plans, 2001-2007.

To describe trends in labor induction, including elective induction, from 2001 to 2007 for six U.S. health plans and to examine the validity of induction measures derived from birth certificate and health plan data. This retrospective cohort study included 339,123 deliveries at 35 weeks' gestation or greater. Linked health plan and birth certificate data provided information about induction, maternal medical conditions, and pregnancy complications. Induction was defined from diagnosis and procedure codes and birth certificate data and considered elective if no accepted indication was coded. We calculated induction prevalence across health plans and years. At four health plans, we reviewed medical records to validate induction measures. Based on electronic data, induction prevalence rose from 28% in 2001 to 32% in 2005, then declined to 29% in 2007. The trend was driven by changes in the prevalence of apparent elective induction, which rose from 11% in 2001 to 14% in 2005 and then declined to 11% in 2007. The trend was similar for subgroups by parity and gestational age. Elective induction prevalence varied considerably across plans. On review of 86 records, 36% of apparent elective inductions identified from electronic data were confirmed as valid. Elective induction appeared to peak in 2005 and then decline. The decrease may reflect quality improvement initiatives or changes in policies, patient or provider attitudes, or coding practices. The low validation rate for measures of elective induction defined from electronic data has important implications for existing quality measures and for research studies examining induction's outcomes.

Variability in calcium/magnesium (CaMg) use for prevention of oxaliplatin-associated neuropathy.

e14644 Background: A retrospective analysis published in June 2004 suggested that patients who received intravenous calcium/magnesium (CaMg) were less likely to experience oxaliplatin-induced neuropathy. Randomized, placebo-controlled trials were developed to test whether CaMg was indeed neuroprotective in patients receiving oxaliplatin. In July 2007, an early Data Monitoring Committee report for one of these trials (CONcePT) described lower response rates in patients randomized to CaMg. However, in March 2008, a revised report negated that claim. In February 2011, publication of a subsequent placebo-controlled study (N04C7) supported that CaMg was possibly neuroprotective, but presentation of a more definitive placebo-controlled trial (N08CB) in June 2013 provided convincing evidence that CaMg does not reduce oxaliplatin-associated neuropathy. Methods: Monthly rates of CaMg the same day as oxaliplatin therapy were assessed using administrative claims data from a large United States (US) health plan. The individuals covered by this health plan, about 88 million between 2003-2013, are geographically dispersed across the US, with the greatest representation in the South and Midwest. We identified receipt of oxaliplatin and CaMg using Healthcare Common Procedure Coding System codes on the claims from 2003 through October 2013. Results: 33,147 patients were included from the health plan. Peaks and valleys in rates of use of CaMg when oxaliplatin was given over the past decade are presented in Table. Conclusions: Substantial swings in CaMg use occurred over time. Prescribing patterns appeared highly sensitive to presentation of study results, suggesting that new scientific evidence was translated into clinical practice in a timely manner. However, some might view these practice changes as premature given that none of the early reports provided conclusive data. Use of CaMg in US claims data. Date % of oxaliplatin doses given with CaMg May 2003 0% May 2007 41% May 2008 5% May 2012 32% Oct 2013 6%

PDB31 - Budget impact analysis of utilizing canagliflozin (CANA) for the treatment of type 2 diabetes mellitus (T2DM) in an United States health plan

PDB31 - Budget impact analysis of utilizing canagliflozin (CANA) for the treatment of type 2 diabetes mellitus (T2DM) in an United States health plan

Colesevelam, Ezetimibe, and Patients With Type 2 Diabetes Mellitus: Characteristics and Clinical Outcomes From a Health Care Database.

Despite the prevalence of therapies available to patients at highest coronary heart disease risk, only a minority of type 2 diabetes mellitus (T2DM) patients reach desired cholesterol treatment levels, with limited data regarding their outcomes. To examine "real-world" effectiveness of initiating treatment with either colesevelam or ezetimibe among individuals with evidence of T2DM and hypercholesterolemia (HCh). Key outcomes included treatment patterns and cardiovascular (CV) events. This retrospective administrative claims-based study utilized medical, pharmacy, and enrollment data linked to laboratory results information from a large United States health plan (January 1, 2006, to March 31, 2011) and included individuals with recorded evidence of T2DM and HCh. The index date was the date of first pharmacy claim for colesevelam or ezetimibe, with cohort assignment based on index medication. Assessments included baseline characteristics, follow-up treatment patterns, and composite CV event, with propensity score matching to correct for sample selection bias. In total, 4231 individuals were identified with evidence of HCh and T2DM (ezetimibe n = 3384; colesevelam n = 847). After matching, the baseline characteristics between cohorts were rendered to be similar. Mean days of persistent medication use was lower with colesevelam compared with ezetimibe (P < 0.001). Compared with ezetimibe, a smaller percentage of individuals in the colesevelam cohort experienced a follow-up composite CV event, and adjusted Cox model results suggested decreased risk (hazard ratio = 0.58; P = 0.004) of a follow-up composite CV event. In this health care database analysis among patients with HCh and T2DM, colesevelam was associated with decreased risk of a composite CV event compared with ezetimibe, despite lower persistence.

Health resource utilization and cost associated with myeloproliferative neoplasms in a large United States health plan

Myelofibrosis (MF), polycythemia vera (PV) and essential thrombocythemia (ET) may lead to bone marrow fibrosis. Because the disease course of ET and PV are long and the disease course of MF may be fatal, healthcare resource utilization (HRU) associated costs of these neoplasms are especially important to understand. We used a large US health insurance claim database to describe the costs of these diseases. Compared to age–gender matched comparisons without myeloproliferative neoplasms (MPN), all aspects of HRU that we examined, including inpatient, outpatient and emergency room visits and pharmacy, as well as overall healthcare expenditures, were significantly higher in patients with MF, PV and ET (e.g. MF total costs = $54 168 vs. $10 203; PV = $14 903 vs. $7913; ET = $29 553 vs. $8026) than in matched comparisons. In order to reduce the burden of illness associated with these diseases, continued efforts in the development of more efficacious treatments for these disorders are needed.

Retrospective claims analysis of best supportive care costs and survival in a US metastatic renal cell population

IntroductionSurvival and best supportive care (BSC) costs for patients with metastatic renal cell carcinoma (mRCC), after stopping therapy, are poorly characterized yet an important aspect of patient care. This study examined survival and costs associated with BSC after one or two lines of therapy (LOTs) for mRCC.MethodsA retrospective cohort analysis used claims data from commercially insured or Medicare Advantage Prescription Drug (MAPD) plan enrollees of a large United States health plan with an index RCC diagnosis (ICD-9-CM 189.0) between January 1, 2007 and June 30, 2010; initiating any of the following therapies 30 days pre-index date through disenrollment from plan: sunitinib, temsirolimus, sorafenib, bevacizumab, everolimus, pazopanib, cytokines. LOT was identified using prescription fill and administration dates. Health care costs represent health plan- plus patient-paid amounts.ResultsThe cohort (n = 274) was 73% male, with a mean age of 63.3 years (SD 11.1), with 80% commercially insured (20% MAPD), and 68% starting BSC following one LOT. Mean BSC duration was longer following one than two LOTs (223 [SD 260], 176 [SD 163] days). Median survival from the start of BSC was similar following one and two LOTs (126 and 118 days). Total BSC costs following one and two LOTs averaged US$50,188 (SD $96,984) and $37,295 (SD $51,102). Monthly costs for BSC following one and two LOTs ($10,151 and $10,566) were not substantially lower than costs while on treatment ($14,621 and $16,957). Inpatient hospital costs represented 47% and 49% following one and two LOTs, with ambulatory costs of approximately 36% following each LOT.ConclusionOur study found similar survival and monthly costs for BSC following either one or two LOTs, with almost half of the cost reflecting inpatient care. Compared to costs on treatment ($14,621 to $16,957), BSC costs can be considerable ($10,151 to $10,566).

OP0277-PC Persistence with Antiosteoporotic Drugs: A Retrospective Population-Based Cohort Study

BackgroundMedication adherence, a synonym for compliance, is the extent to which patients take medication at the prescribed interval and dose. Persistence is defined as the period between the start and...

Risk of hospitalized rhabdomyolysis associated with lipid-lowering drugs in a real-world clinical setting

The occurrence of low rates of rhabdomyolysis among patients receiving lipid-lowering drugs (LLDs) in randomized clinical trials may be elucidated with population-based studies. To determine the risk of hospitalized rhabdomyolysis associated with LLD therapy. This observational study used claims data from 9 million members of five United States health plans to identify patients (≥18 years) who received >2 statin and nonstatin LLDs during July 2000 to December 2004. Inpatient International Classification of Diseases, Ninth Revision, Clinical Modification, codes for rhabdomyolysis (791.3, 728.89, and 728.88) were observed during the follow-up period; cases were confirmed with patients' medical records. Rhabdomyolysis events were reported per 10,000 person-years of LLD exposure; multivariate analysis was conducted. The study cohort (N = 473,343) received 490,988 and 11,624 person-years of LLD, and combination therapy, respectively. Medical charts were obtained for 104 of 144 eligible patients with rhabdomyolysis claims; 42 cases were confirmed. With atorvastatin as reference, rhabdomyolysis rates (95% confidence interval) were greatest for cerivastatin, 8.4 (2.3-21.7); no difference among available statins was observed. Rates for other LLD monotherapies were: niacin, 2.1 (0.3-7.7), ezetimibe, 2.1 (0.3-7.8), fenofibrate, 0 (0-1.7), and gemfibrozil, 2.0 (0.5-5.2). Multivariate analysis showed only cerivastatin with a significantly greater risk of rhabdomyolysis (odds ratio 4.74, 95% confidence interval 1.1-21.2, P = .041) versus atorvastatin among the statins. Combination therapies had increased rhabdomyolysis risk (OR 7.1, 1.6-31.6, P = .010) versus LLDs alone. The risk of habdomyolysis among hospitalized patients receiving statins was low; no difference among the available statins was evident. Further data are needed to establish the risk profile but current findings already offer guidance to physicians.

Psoriasis treatment patterns with etanercept and adalimumab in a United States health plan population

Background: A retrospective study utilizing administrative claims from a US commercial health plan was performed to examine etanercept and adalimumab treatment patterns among patients with psoriasis (PSO). Methods: Biologic-naïve PSO patients initiating etanercept or adalimumab therapy between 18 January 2008 and 31 December 2008 were identified. Patients continuously enrolled in the health plan for 6 months before and ≥12 months after therapy initiation were followed until disenrollment from the plan or 31 December 2009. Persistence was defined as continuous use of index TNF blocker without a gap in therapy ≥60 days. Patients with gaps in index therapy ≥60 days were classified as discontinuing, switching, or restarting the index therapy. Results: In total, 497 patients initiated etanercept and 330 the adalimumab therapy. Mean age for both groups was 43 years. Approximately 40–42% of patients were persistent on their index TNF blocker for 1 year. Among patients with a ≥60-day gap in therapy, discontinuation without restart or switch occurred in 37% of etanercept and 45% of adalimumab patients (p = 0.04). Differences in therapy restart or switching between the groups were not statistically significant. Conclusions:TNF-blocker therapy persistence is low among PSO patients in this health plan. More than one-third of patients restarted their index TNF blocker after a gap in therapy.

PSS29 Real-World Characteristics and Severity Assessment of United States Health Plan Members Treated with Ustekinumab for Moderate-to-Severe Psoriasis

PSS29 Real-World Characteristics and Severity Assessment of United States Health Plan Members Treated with Ustekinumab for Moderate-to-Severe Psoriasis

Health plan budget impact of introducing everolimus for treatment of advanced progressive pancreatic neuroendocrine tumors in the United States.

229 Background: Everolimus was recently approved for treatment of patients with advanced progressive pancreatic neuroendocrine tumors (pNET) based on a clinically meaningful and statistically significant increase in progression-free survival. To understand the financial impact of introducing everolimus into the market, a model was developed to simulate a typical health plan in the United States (US). Methods: A cross-sectional budget impact model with a one-year time horizon was developed. Disease prevalence and drug usage data were derived from a real-world data source. Costs for anti-tumor therapy and underlying care included concomitant medications, physician visits, treatment infusion, surgical procedures, tests, hospitalizations, and adverse event (AE) management. Resource utilization rates were derived from a physician survey-based resource utilization study; treatment-related AE rates were taken from secondary literature. Drug costs were Wholesale Acquisition Costs (2011). Medical procedure and hospitalization costs were based on Medicare fee schedules for current procedure terminology and diagnosis-related group codes. AE costs were obtained from the literature. The model assessed the initial annual budget impact, under the assumption that everolimus and sunitinib, another recently approved targeted therapy, replace most currently available chemotherapy treatments. Results: Based on published prevalence rates, for a health plan with 1,000,000 members, an annual count of 12 advanced pNET patients was projected. The model estimated a $22,657 budget increase ($0.0019 per member per month or $0.0227 per member per year [PMPY]). The largest percent increase in budget was associated with anti-tumor treatments; infusions and surgical procedures represented the largest percent decrease in costs. Scenario analyses indicated that increasing the percentage of everolimus use (by reducing that of sunitinib) resulted in a decrease in the overall budget compared to baseline. Conclusions: Our model predicts that treating pNET patients with everolimus will have a minimal budget impact on US health plans, amounting to about $0.027 PMPY.

Medication adherence and fracture risk among patients on bisphosphonate therapy in a large United States health plan

The association between bisphosphonate adherence in the first 12months after therapy initiation and subsequent fracture risk was examined. Patients were identified from a large, commercially-insured population with integrated pharmacy and medical claims. Eligible patients were aged ≥45years, were new to osteoporosis therapy (no osteoporosis medication claims in prior year) with first (index) bisphosphonate claim between 1/1/2005 and 4/30/2008, and had continuous insurance coverage for ≥12months pre- and post-index. Patients with fracture claims≤12-months post-index were excluded. Adherence was assessed using the medication possession ratio (MPR) over 12-months post-index (i.e., sum of days' supply dispensed divided by 365days). Patients with a MPR>0.8 were considered adherent. The follow-up period to assess incident fracture began at month 13. The analysis included 33,558 new bisphosphonate users with mean age (SD) 59.5 (9.3) years; 94.0% were female. Median MPR at 12months was 0.61 for alendronate and risedronate; 0.58 for ibandronate. Proportionally more nonfracture patients (39.3%) had a MPR>0.8 compared with fracture patients (34.9%, p<0.001). In multivariate modeling of bisphosphonate users' experience, those with a MPR>0.8 had a 14% lower risk of subsequent fracture than those with MPR<0.5, after controlling for demographics, insurance type, select comorbidities, and other potential confounders (p=0.0459). In a large, commercially-insured population, suboptimal adherence with bisphosphonate treatment was associated with increased fracture risk even after controlling for potential confounders.