Cost-effectiveness of apixaban vs. current standard of care for stroke prevention in patients with atrial fibrillation.

AIMS: Warfarin, a vitamin K antagonist (VKA), has been the standard of care for stroke prevention in patients with atrial fibrillation (AF). Aspirin is recommended for low-risk patients and those unsuitable for warfarin. Apixaban is an oral anticoagulant that has demonstrated better efficacy than warfarin and aspirin in the ARISTOTLE and AVERROES studies, respectively, and causes less bleeding than warfarin. We evaluated the potential cost-effectiveness of apixaban against warfarin and aspirin from the perspective of the UK payer perspective. RESULTS AND METHODS: A lifetime Markov model was developed to evaluate the pharmacoeconomic impact of apixaban compared with warfarin and aspirin in VKA suitable and VKA unsuitable patients, respectively. Clinical events considered in the model include ischaemic stroke, haemorrhagic stroke, intracranial haemorrhage, other major bleed, clinically relevant non-major bleed, myocardial infarction, cardiovascular hospitalization and treatment discontinuations; data from the ARISTOTLE and AVERROES trials and published mortality rates and event-related utility rates were used in the model. Apixaban was projected to increase life expectancy and quality-adjusted life years (QALYs) compared with warfarin and aspirin. These gains were expected to be achieved at a drug acquisition-related cost increase over lifetime. The estimated incremental cost-effectiveness ratio was £11 909 and £7196 per QALY gained with apixaban compared with warfarin and aspirin, respectively. Sensitivity analyses indicated that results were robust to a wide range of inputs. CONCLUSIONS: Based on randomized trial data, apixaban is a cost-effective alternative to warfarin and aspirin, in VKA suitable and VKA unsuitable patients with AF, respectively.

Evaluation of cardiovascular morbidity associated with adherence to atorvastatin therapy.

Long-term adherence to statins is poor. We assessed the relationship between cardiovascular (CV) risk and atorvastatin adherence in primary- and secondary-prevention patients, adjusting for healthy-adherer bias by incorporating preventive service use into the model. Medical and pharmacy claims from employee-based plans from 2002 to 2008 were analyzed for patients who initiated atorvastatin in 2003-2004. Adherent patients were defined as having ≥60% of days covered in the year after atorvastatin initiation and were required to have pill coverage in months 10-12. CV events were identified as hospitalizations with a primary CV diagnosis and assessed from month 13 after atorvastatin initiation until the end of follow-up (≤36 months). Cox proportional hazards models were used to examine the association between atorvastatin adherence and CV event risk, adjusting for covariates including preventive service use. The study included 94,287 atorvastatin users (79,010 primary- and 15,277 secondary-prevention patients). In both populations, nearly one-half of the patients discontinued atorvastatin after 1 year. During follow-up, ~2% of primary-prevention and ~9% of secondary-prevention patients experienced CV events. After adjusting for covariates, adherent patients in the primary-prevention population had a significantly lower risk of CV events compared with nonadherent patients (hazard ratio, 0.82; 95% confidence interval, 0.74-0.91). In the secondary-prevention population, adherence to atorvastatin was also associated with lower CV risk (hazard ratio, 0.74; 95% confidence interval, 0.66-0.82). Atorvastatin discontinuation rates were high 1 year after treatment initiation. Patients who adhered to atorvastatin treatment were at lower CV risk. Quality-of-care interventions should target improvements to therapy persistence.

New diagnosis of hypertension among celecoxib and nonselective NSAID users: a population-based cohort study.

BACKGROUND: The use of nonsteroidal antiinflammatory drugs (NSAIDs) has been associated with increased blood pressure and hypertension. However, less is known about how the risk of hypertension is associated with cyclooxygenase 2 selective inhibitors (coxibs), especially celecoxib, the only coxib remaining on the market. OBJECTIVE: To compare the risk of incident hypertension associated with the use of celecoxib and nonselective (NS) NSAIDs. METHODS: A cohort study was conducted using secondary data from the GE Centricity Electronic Medical Record database, which contains millions of patient records seen by thousands of physicians across the US. The index date was defined as the date of the first NSNSAID or celecoxib prescription between January 1, 1999, and June 30, 2004. Patients were included if they were aged 18 years or older and were enrolled for at least 365 days prior to the index date. Excluded were patients who had any prior diagnosis of hypertension or pregnancy-related hypertension during the pre- or postindex date period. Also excluded were patients who had any prior prescription for antihypertensive drugs, coxibs, or NSNSAIDs. After applying inclusion/exclusion criteria, each celecoxib user was matched to 2 NSNSAID users by sex, age (+/-5 y), propensity score (within 0.2 SD), and number of unique drugs (+/-20%). Descriptive and survival analyses were conducted. RESULTS: The final sample consisted of 51,444 patients, of whom 17,148 were on celecoxib and 34,296 were on NSNSAIDs. Relative to NSNSAID users, patients on celecoxib had a similar rate of postexposure hypertension incidence (HR 1.013; 95% CI 0.862 to 1.190). CONCLUSIONS: Results from a population-based cohort analysis of electronic medical records did not show any difference in the hazard rates of incident hypertension between celecoxib and NSNSAID users.

Impact of statin copayments on adherence and medical care utilization and expenditures.

OBJECTIVE: To examine the effects of statin cost-sharing (ie, copayments, coinsurance) on adherence to statin medications and the impact of adherence on healthcare utilization and spending. STUDY DESIGN: Retrospective, observational study of statin users receiving health benefits or supplemental coverage from employer-sponsored health plans. METHODS: Medical and pharmacy claims were selected from the Medstat MarketScan database for patients who were continuously enrolled from 2000 through 2003. Two-stage residual inclusion models were estimated. The first stage modeled adherence to statins, which was derived from the medication possession ratio, and represented the percentage of days on therapy in an 18-month time frame, July 2001 through December 2002. In the second stage, generalized linear models were used to estimate 2003 utilization and expenditures. Separate estimates were produced for new statin users (n = 24 113) and continuing statin users (n = 93 253). RESULTS: Lower statin copayments were associated with higher levels of statin adherence. In percentage terms, when holding all other variables at their mean value, a $10 increase in copayment resulted in a 1.8 percentage point reduction in the probability of adherence for new users and a 3 percentage point reduction in the probability of adherence for continuing users. For continuing users adherent to statins, total costs did not change, but fewer negative events (emergency department visits, hospitalizations, and coronary heart disease-related hospitalizations) occurred. CONCLUSIONS: Policy makers and plan managers should consider interventions that improve adherence to statins, such as lower copayments.

Opioid expenditures and utilization in the Medicaid system.

Data from the U.S. Centers for Medicare and Medicaid Services' Medicaid database were analyzed to define recent trends in the use and cost of opioid medications United States as a whole and in seven states between 1998 and 2003. Over this time period, total Medicaid prescriptions for opioids have nearly doubled and, in nominal terms, expenditures have nearly tripled. The morphine derivatives account for the largest share of these increases. By 2003, opioids represented about a four percent share of all Medicaid prescription drugs. Variations among the states in per-enrollee spending on opioids are substantial. The appropriateness of these variations is unknown.

Reverse Engineering and Evaluation of Prediction Models for Progression to Type 2 Diabetes: An Application of Machine Learning Using Electronic Health Records.

BACKGROUND: Application of novel machine learning approaches to electronic health record (EHR) data could provide valuable insights into disease processes. We utilized this approach to build predictive models for progression to prediabetes and type 2 diabetes (T2D). METHODS: Using a novel analytical platform (Reverse Engineering and Forward Simulation [REFS]), we built prediction model ensembles for progression to prediabetes or T2D from an aggregated EHR data sample. REFS relies on a Bayesian scoring algorithm to explore a wide model space, and outputs a distribution of risk estimates from an ensemble of prediction models. We retrospectively followed 24 331 adults for transitions to prediabetes or T2D, 2007-2012. Accuracy of prediction models was assessed using an area under the curve (AUC) statistic, and validated in an independent data set. RESULTS: Our primary ensemble of models accurately predicted progression to T2D (AUC = 0.76), and was validated out of sample (AUC = 0.78). Models of progression to T2D consisted primarily of established risk factors (blood glucose, blood pressure, triglycerides, hypertension, lipid disorders, socioeconomic factors), whereas models of progression to prediabetes included novel factors (high-density lipoprotein, alanine aminotransferase, C-reactive protein, body temperature; AUC = 0.70). CONCLUSIONS: We constructed accurate prediction models from EHR data using a hypothesis-free machine learning approach. Identification of established risk factors for T2D serves as proof of concept for this analytical approach, while novel factors selected by REFS represent emerging areas of T2D research. This methodology has potentially valuable downstream applications to personalized medicine and clinical research.

LDL-C goal attainment in patients who remain on atorvastatin or switch to equivalent or non-equivalent doses of simvastatin: a retrospective matched cohort study in clinical practice.

BACKGROUND: As clinical trials have shown the benefits of more intensive cholesterol control, treatment targets for low-density lipoprotein cholesterol (LDL-C) have decreased progressively. At the same time, physicians have been encouraged to contain costs by prescribing cheaper, generic statins for cholesterol management. To determine how these possibly conflicting goals are managed in clinical practice, we examined LDL-C control in patients switched from a potent, branded statin (atorvastatin) to a less potent, generic statin (simvastatin). METHODS: Patients who switched from atorvastatin to simvastatin between July 2006 and January 2008 were retrospectively identified from a US medical and pharmacy claims database, and matched with controls remaining on atorvastatin. Outcomes measured were the number of switched patients receiving a simvastatin milligram dose>or=2 times their previous atorvastatin dose, changes in LDL-C levels, and percentage of patients achieving recommended LDL-C targets. All study variables were analyzed descriptively. RESULTS: After applying exclusion and inclusion criteria, 1048 patients who switched from atorvastatin to simvastatin and 1048 matched controls who remained on atorvastatin were included. Among the switchers, 379 (36%) received an inappropriately low dose of simvastatin (<2 times atorvastatin dose). In patients remaining on atorvastatin, mean LDL-C decreased from 105.7 mg/dL to 102.3 mg/dL after 44 weeks, whereas in switched patients, LDL-C remained similar, at 105.9 mg/dL on atorvastatin and 105.8 mg/dL on simvastatin. Before switching, when all patients were receiving atorvastatin, 67.4% of switchers and 69.9% of controls achieved recommended LDL-C targets. After switching, significantly fewer switchers than controls met LDL-C targets (69.1% vs 74.6%; P=0.005). However, among patients who switched to an equivalent dose of simvastatin (>or=2 times prior atorvastatin dose), similar proportions met LDL-C targets (72.8% vs 74.6% of controls; P=0.402), whereas among patients who switched to inappropriate non-equivalent dose of simvastatin, a significantly lower proportion met LDL-C targets (62.5% vs 74.6% of controls; P=0.001). CONCLUSIONS: Continuing atorvastatin was associated with lower LDL-C levels and better LDL-C target attainment compared with switching to simvastatin. Patients switched to an equivalent simvastatin dose had lower LDL-C levels and were more likely to achieve LDL-C targets than patients switched to a non-equivalent dose, suggesting physicians must consider dosage equivalence when switching statins, and should measure LDL-C and titrate statins as necessary to achieve LDL-C control.

Valdecoxib is associated with improved dyspepsia-related health compared with nonspecific NSAIDs in patients with osteoarthritis or rheumatoid arthritis.

OBJECTIVES: Dyspepsia and related gastrointestinal (GI) symptoms are commonly reported by patients taking nonspecific nonsteroidal anti-inflammatory drugs (NSAIDs) and significantly impact treatment effectiveness, cost, and quality of life. This study sought to evaluate dyspepsia-related health in osteoarthritis (OA) and rheumatoid arthritis (RA) patients taking valdecoxib compared with patients taking nonspecific NSAIDs. METHODS: Analysis of two separate, double-blind, placebo-controlled studies: one in RA patients randomized to placebo, valdecoxib (10 and 20 mg once daily [o.d.]) and naproxen (500 mg twice daily [b.i.d.]); one in OA patients randomized to placebo, valdecoxib (10 and 20 mg o.d.), diclofenac (75 mg b.i.d.), or ibuprofen (800 mg three times daily [t.i.d.]). Study population comprised patients with RA in flare or clinically documented OA who required chronic symptomatic treatment with NSAIDs/analgesics. Dyspepsia-related health was evaluated at baseline and weeks 2, 6, and 12 (or early termination) using the validated Severity of Dyspepsia Assessment (SODA) questionnaire. This patient self-report tool consists of scales for evaluating dyspepsia pain intensity, nonpain symptoms, and satisfaction. Analysis was based on the intent-to-treat population with the last observation carried forward. RESULTS: Valdecoxib was significantly better at endpoint than standard doses of naproxen, diclofenac, and ibuprofen for pain intensity scores (p < 0.05), and provided significantly improved nonpain symptom and satisfaction scores compared with naproxen for patients with RA (p < 0.05). For RA patients, the difference between valdecoxib and naproxen pain intensity scores were clinically meaningful; at all the time points, significantly fewer patients receiving valdecoxib reported severe dyspepsia pain intensity increases (>/=10 points) than those receiving naproxen. At 12 wk, fewer patients receiving valdecoxib reported severe dyspepsia pain intensity increases versus those receiving ibuprofen and diclofenac. CONCLUSIONS: The GI tolerability of valdecoxib is superior to that of nonspecific NSAIDs, and therefore can potentially have a favorable impact on patient quality of life.