Publications

OBJECTIVE: Underuse of clozapine and overuse of antipsychotic polypharmacy are both indicators of poor quality of care. This study examined variation in prescribing clozapine and antipsychotic polypharmacy across providers, as well as factors associated with these practices.

METHODS: Using 2010-2012 Pennsylvania Medicaid data, prescribers were identified if they wrote antipsychotic prescriptions for ten or more nonelderly adult patients with schizophrenia annually. Generalized linear mixed models with a binomial distribution and a logit link were used to examine prescriber-level annual percentages of patients with clozapine use and with long-term (≥90 days) antipsychotic polypharmacy and associated characteristics of prescribers' patient caseloads, prescriber characteristics, and Medicaid payer (fee-for-service versus managed care plans).

RESULTS: The study cohort included 645 prescribers in 2010, 632 in 2011, and 650 in 2012. In 2012, the mean prescriber-level annual percentage of patients with any clozapine use was 7% (range 0%-89%), and the mean percentage of patients with any long-term antipsychotic polypharmacy was 7% (range 0%-45%) (similar rates were found during 2010-2012). Prescribers with high prescription volume, a smaller percentage of patients from racial or ethnic minority groups, and a larger percentage of patients eligible for Supplemental Security Income were more likely to use both clozapine and antipsychotic polypharmacy for treating schizophrenia. Prescriber specialty and Medicaid payer were also associated with prescribers' practices.

CONCLUSIONS: Considerable variation was found in clozapine and antipsychotic polypharmacy practices across prescribers in their treatment of schizophrenia. Targeting efforts to selected prescribers holds promise as an approach to promote evidence-based antipsychotic prescribing.

INTRODUCTION: Expanding Medicaid coverage to low-income adults may have increased smoking cessation through improved access to evidence-based treatments. Our study sought to determine if states' decisions to expand Medicaid increased recent smoking cessation.

METHODS: Using pooled cross-sectional data from the Behavioral Risk Factor Surveillance Survey for the years 2011-2015, we examined the association between state Medicaid coverage and the probability of recent smoking cessation among low-income adults without dependent children who were current or former smokers (n=36,083). We used difference-in-differences estimation to examine the effects of Medicaid coverage on smoking cessation, comparing low-income adult smokers in states with Medicaid coverage to comparable adults in states without Medicaid coverage, with ages 18-64 years to those ages 65 years and above. Analyses were conducted for the full sample and stratified by sex.

RESULTS: Residence in a state with Medicaid coverage among low-income adult smokers ages 18-64 years was associated with an increase in recent smoking cessation of 2.1 percentage points (95% confidence interval, 0.25-3.9). In the comparison group of individuals ages 65 years and above, residence in a state with Medicaid coverage expansion was not associated with a change in recent smoking cessation (-0.1 percentage point, 95% confidence interval, -2.1 to 1.8). Similar increases in smoking cessation among those ages 18-64 years were estimated for females and males (1.9 and 2.2 percentage point, respectively).

CONCLUSION: Findings are consistent with the hypothesis that Medicaid coverage expansions may have increased smoking cessation among low-income adults without dependent children via greater access to preventive health care services, including evidence-based smoking cessation services.

OBJECTIVE: Second-generation antipsychotics increase the risk of diabetes and other metabolic conditions among individuals with schizophrenia. Although metabolic testing is recommended to reduce this risk, low testing rates have prompted concerns about negative health consequences and downstream medical costs. This study simulated the effect of increasing metabolic testing rates on ten-year prevalence rates of prediabetes and diabetes (diabetes conditions) and their associated health care costs.

METHODS: A microsimulation model (N=21,491 beneficiaries) with a ten-year time horizon was used to quantify the impacts of policies that increased annual testing rates in a Medicaid population with schizophrenia. Data sources included California Medicaid data, National Health and Nutrition Examination Survey data, and the literature. In the model, metabolic testing increased diagnosis of diabetes conditions and diagnosis prompted prescribers to switch patients to lower-risk antipsychotics. Key inputs included observed diagnoses, prescribing rates, annual testing rates, imputed rates of undiagnosed diabetes conditions, and literature-based estimates of policy effectiveness.

RESULTS: Compared with 2009 annual testing rates, ten-year outcomes for policies that achieved universal testing reduced exposure to higher-risk antipsychotics by 14%, time to diabetes diagnosis by 57%, and diabetes prevalence by .6%. These policies were associated with higher spending because of testing and earlier treatment.

CONCLUSIONS: The model showed that policies promoting metabolic testing provided an effective approach to improve the safety of second-generation antipsychotic prescribing in a Medicaid population with schizophrenia; however, the policies led to additional costs at ten years. Simulation studies are a useful source of information on the potential impacts of these policies.

Optimism and cynical hostility are associated with health behaviors and health outcomes, including morbidity and mortality. This analysis assesses their association with longitudinal vigorous physical activity (PA) in postmenopausal women of the Women's Health Initiative (WHI). Subjects include 73,485 women nationwide without history of cancer or cardiovascular disease (CVD), and no missing baseline optimism, cynical hostility, or PA data. The Life Orientation Test-Revised Scale measured optimism. A Cook Medley questionnaire subscale measured cynical hostility. Scale scores were divided into quartiles. Vigorous PA three times or more per week was assessed via self-report at study baseline (1994-1998) and through follow-up year 6. Descriptive analysis mapped lifetime trajectories of vigorous PA (recalled at ages 18, 25, 50; prospectively assessed at baseline, and 3 and 6 years later). Hierarchical generalized linear mixed models examined the prospective association between optimism, cynical hostility, and vigorous PA over 6 years. Models adjusted for baseline sociodemographic variables, psychosocial characteristics, and health conditions and behaviors. Vigorous PA rates were highest for most optimistic women, but fell for all women by approximately 60% between age 50 and study baseline. In adjusted models from baseline through year 6, most vs. least optimistic women were 15% more likely to exercise vigorously (p < 0.001). Cynical hostility was not associated with lower odds of longitudinal vigorous PA after adjustment. Results did not differ by race/ethnicity or socioeconomic status. Higher optimism is associated with maintaining vigorous PA over time in post-menopausal women, and may protect women's health over the lifespan.

In 2009, Blue Cross-Blue Shield of Massachusetts (BCBSMA) implemented the alternative quality contract (AQC), which pays provider organizations a global payment for all services used by enrollees. BCBSMA claims for 2006-2011 were used to compare youths enrolled in provider organizations participating in the AQC (7,407 person-years [PYs]) with those not participating (45,398 PYs). Difference-in-differences models estimated changes in mental health and substance abuse treatment service utilization and spending attributable to the AQC. The AQC was associated with small increases in the probability of any outpatient visits and in the probability and number of medication management visits among children with attention-deficit hyperactivity disorder (ADHD). Spending did not change, and there was no evidence of reductions in service utilization or spending for children with ADHD in the first three years of AQC implementation.

PURPOSE: The findings from the observational studies comparing the effectiveness and safety of non-vitamin K antagonist oral anticoagulants (NOACs) versus vitamin K antagonists (VKAs) for atrial fibrillation (AF) and venous thromboembolism (VTE) are inconsistent. We conducted separate meta-analyses examining the efficacy/effectiveness and safety of NOACs versus VKAs by disease (AF vs VTE), study design (randomized controlled trials [RCTs] vs observational studies), and NOAC (dabigatran, rivaroxaban, apixaban, and edoxaban).

METHODS: The main data sources included PubMed/MEDLINE, EMBASE, Web of Science, CINAHL, and Scopus from January 1, 2005, to February 15, 2016. We searched for Phase III RCTs and observational studies comparing NOACs versus VKAs. The primary outcomes were stroke/systemic embolism (SE) for AF; recurrent VTE/fatal pulmonary embolism (PE) for VTE; and major bleeding for both conditions. Secondary outcomes included stroke and myocardial infarction (MI) for AF, recurrent deep vein thrombosis (DVT)/PE for VTE, and mortality, intracranial hemorrhage (ICH), and gastrointestinal bleeding for both conditions. Pooled hazard ratios (HRs) were reported by using inverse variance-weighted random effects models.

FINDINGS: A total of 13 RCTs and 27 observational studies (AF, n = 32; VTE, n = 8) were included. For AF, dabigatran and VKAs were comparable for stroke/SE risk in 1 RCT (HR, 0.77 [95% CI, 0.57-1.03]) and 6 observational studies (HR, 1.03 [95% CI, 0.83-1.27]). Rivaroxaban had a 20% decreased risk of stroke/SE in 3 RCTs (HR, 0.80 [95% CI, 0.67-0.95]) compared with VKA, but the effect was nonsignificant in 3 observational studies (HR, 0.78 [95% CI, 0.59-1.04]). Apixaban decreased stroke/systemic embolism risk (HR, 0.79 [95% CI, 0.66-0.95]) compared with VKA in 1 RCT, but edoxaban was comparable to VKA (HR, 0.99 [95% CI, 0.77-1.28]) in 1 RCT (no observational studies available for apixaban/edoxaban). Dabigatran, apixaban, and edoxaban decreased the risk of hemorrhagic stroke, mortality, major bleeding, and ICH by 10% to 71% compared with VKAs but not rivaroxaban. For VTE, NOACs and VKAs were comparable for recurrent VTE/fatal PE/DVT/PE risk in 7 RCTs and 1 observational study. The 7 RCTs demonstrated a 32% to 69% decreased risk of major bleeding for dabigatran, rivaroxaban, and apixaban compared with VKAs. No difference was shown in 1 rivaroxaban observational study (HR, 0.77 [95% CI, 0.40-1.49]) and 1 edoxaban RCT (HR, 0.84 [95% CI, 0.59-1.20]). Except for dabigatran, the NOACs had a 61% to 86% decreased risk of ICH and gastrointestinal bleeding.

IMPLICATIONS: Overall, NOACs were comparable or superior to VKAs. Although no observational studies are currently available for apixaban/edoxaban, a few notable inconsistencies exist for dabigatran (ischemic stroke, MI) and rivaroxaban (stroke/SE, major bleeding in VTE) between RCTs and observational studies. Individualizing NOAC/VKA therapy based on benefit/safety profiles and patient characteristics is suggested.

OBJECTIVES: The US opioid medication epidemic has resulted in serious health consequences for patients. Formulary management tools adopted by payers, specifically prior authorization (PA) policies, may lower the rates of opioid medication abuse and overdose. We compared rates of opioid abuse and overdose among enrollees in plans that varied in their use of PA from "High PA" (ie, required PA for 17 to 74 opioids), with "Low PA" (ie, required PA for 1 opioid), and "No PA" policies for opioid medications.

STUDY DESIGN: Retrospective cohort study of patients initiating opioid treatment in Pennsylvania Medicaid from 2010 to 2012.

METHODS: Generalized linear models with generalized estimating equations were employed to assess the relationships between the presence of PA policies and opioid medication abuse and overdose, as measured in Medicaid claims data, adjusting for demographics, comorbid health conditions, benzodiazepine/muscle relaxant use, and emergency department use.

RESULTS: The study cohort included 297,634 enrollees with a total of 382,828 opioid treatment episodes. Compared with plans with No PA, enrollees in High PA (adjusted rate ratio [ARR], 0.89; 95% confidence interval [CI], 0.85-0.93; P <.001) and Low PA plans (ARR, 0.93; 95% CI, 0.87-1.00; P = .04) had lower rates of abuse. Enrollees in the Low PA plan had a lower rate of overdose than those within plans with No PA (ARR, 0.75; 95% CI, 0.59-0.95; P = .02). High PA plan enrollees were also less likely than No PA enrollees to experience an overdose, but this association was not statistically significant (ARR, 0.88; 95% CI, 0.76-1.02; P = .08).

CONCLUSIONS: Enrollees within Medicaid plans that utilize PA policies appear to have lower rates of abuse and overdose following initiation of opioid medication treatment.

BACKGROUND: Health systems may play an important role in identification of patients at-risk of opioid medication overdose. However, standard measures for identifying overdose risk in administrative data do not exist.

OBJECTIVE: Examine the association between opioid medication overdose and 2 validated measures of nonmedical use of prescription opioids within claims data.

RESEARCH DESIGN: A longitudinal retrospective cohort study that estimated associations between overdose and nonmedical use.

SUBJECTS: Adult Pennsylvania Medicaid program 2007-2012 patients initiating opioid treatment who were: nondual eligible, without cancer diagnosis, and not in long-term care facilities or receiving hospice.

MEASURES: Overdose (International Classification of Disease, ninth edition, prescription opioid poisonings codes), opioid abuse (opioid use disorder diagnosis while possessing an opioid prescription), opioid misuse (a composite indicator of number of opioid prescribers, number of pharmacies, and days supplied), and dose exposure during opioid treatment episodes.

RESULTS: A total of 372,347 Medicaid enrollees with 583,013 new opioid treatment episodes were included in the cohort. Opioid overdose was higher among those with abuse (1.5%) compared with those without (0.2%, P<0.001). Overdose was higher among those with probable (1.8%) and possible (0.9%) misuse compared with those without (0.2%, P<0.001). Abuse [adjusted rate ratio (ARR), 1.52; 95% confidence interval (CI), 1.10-2.10), probable misuse (ARR, 1.98; 95% CI, 1.46-2.67), and possible misuse (ARR, 1.76; 95% CI, 1.48-2.09) were associated with significantly more events of opioid medication overdose compared with those without.

CONCLUSIONS: Claims-based measures can be used by health systems to identify individuals at-risk of overdose who can be targeted for restrictions on opioid prescribing, dispensing, or referral to treatment.