Use of Pooled State Administrative Data for Mental Health Services Research.

State systems are a rich, albeit challenging, laboratory for policy-relevant services research studies. State mental health authorities routinely devote resources to collect data for state planning and reporting purposes. However, these data are rarely used in cross-state comparisons to inform state or federal policy development. In 2008, in response to key recommendations from the National Institute of Mental Health (NIMH) Advisory Council's "The Road Ahead: Research Partnership to Transform Services," (http://www.nimh.nih.gov/about/advisory-boards-and-groups/namhc/reports/road-ahead.pdf), NIMH issued a request for applications (RFA) to support studies on the impact of state policy changes on access, cost, quality and outcomes of care for individuals with mental disorders. The purpose of the RFA was to bridge the divide between research and policy by encouraging research that used state administrative data across states, and to address significant state-defined health policy initiatives. Five projects involving eight states were selected through peer review for funding. Projects began in 2009 and were funded for 3 years. This report provides a brief description of the five projects, followed by an analysis of the impact, challenges, and lessons learned from these policy-partnered studies. We conclude by offering suggestions on ways to use state administrative data for informing state health policies, which is especially timely given national and state changes in the structure and financing of healthcare.

A framework for describing health care delivery organizations and systems.

Describing, evaluating, and conducting research on the questions raised by comparative effectiveness research and characterizing care delivery organizations of all kinds, from independent individual provider units to large integrated health systems, has become imperative. Recognizing this challenge, the Delivery Systems Committee, a subgroup of the Agency for Healthcare Research and Quality's Effective Health Care Stakeholders Group, which represents a wide diversity of perspectives on health care, created a draft framework with domains and elements that may be useful in characterizing various sizes and types of care delivery organizations and may contribute to key outcomes of interest. The framework may serve as the door to further studies in areas in which clear definitions and descriptions are lacking.

Advancing geriatrics research, education, and practice: policy challenges after the great recession.

The series of articles in this JGIM issue provides a number of policy-relevant recommendations for advancing geriatrics research, education and practice. Despite the unprecedented pressure to reduce state and federal spending, policymakers must concurrently address the challenges of a growing population of older individuals with increasingly complex health care problems. Thus, there may be opportunities to advance this agenda in creative ways. For example, without new spending, federal research agencies can make changes to encourage needed new directions in aging research, and the ACA provides new funding opportunities such as the Patient Centered Outcomes Research Institute. States and the federal government have an increasing need for the health professions workforce to have collaborative care skills and geriatrics clinical competencies, and are finding ways to invest in relevant initiatives. On the clinical program side, state and federal governments are initiating programs to promote delivery system changes that improve the care of older adults. Nonetheless, in the face of the policy challenges that have persisted after the "great recession," academic geriatrics and general internal medicine will need to join forces with public and private interests to secure the resources needed to advance this ambitious agenda for geriatrics research, education and practice.

Characteristics and trends of low-dose quetiapine use in two western state Medicaid programs.

PURPOSE: Medicaid programs are concerned about inappropriate, potentially hazardous, and costly off-label use of second-generation antipsychotics (SGAs). Several states are exploring policies aimed at managing low-dose quetiapine, commonly prescribed for off-label conditions. This study aimed to characterize longitudinal trends and patient characteristics associated with low-dose quetiapine in two state Medicaid programs. We further aimed to quantify changes in the use of quetiapine associated with a legal settlement that curtailed off-label promotion of this product. METHODS: Using administrative data from two state Medicaid programs, Oregon and Colorado, we identified SGA initiators and determined patient level factors associated with receipt of low-dose SGAs. We evaluated changes in low-dose quetiapine initiation during and after a period in which quetiapine was being promoted illegally for off-label purposes. RESULTS: We identified 14,763 new SGA starts during the study period. Low-dose (versus therapeutic dose) SGA use was common in both states, representing 53% to 56% of initiators. Quetiapine was the most commonly used SGA in both states and both dose ranges. Diagnoses of schizophrenia, bipolar disorder, posttraumatic stress disorder, anxiety disorder, and use of newer sedative hypnotics were associated with lower likelihood of initiating low-dose quetiapine. Initiation of low-dose quetiapine as a proportion of all SGA initiation and of all quetiapine initiation significantly declined in Oregon following suspension of off-label promotional activities. CONCLUSIONS: Low-dose SGA and specifically low-dose quetiapine use remains common. Medicaid programs must set policies carefully to maximize the net safety of prescription use while optimizing disease management considering the potential for substitution effects.

Potential barriers associated with increased prevalence of perforated appendicitis in Colorado's pediatric Medicaid population.

OBJECTIVE: We sought to identify barriers and delays in care associated with the increased prevalence of perforated appendicitis among Colorado's pediatric Medicaid population. METHODS: We conducted a retrospective cohort study of all cases of pediatric appendicitis, which had Colorado Medicaid from 2007 to 2008 using descriptive statistics, bivariate analysis, and multivariable logistic regression. RESULTS: Of the 479 appendicitis cases, 42.6% were perforated. In both the bivariate and multivariate analysis, perforated cases did not significantly differ from nonperforated cases with respect to sex, rurality of residence, or race with the exception of black race in the multivariate model. Perforated cases were more likely to be younger, have been enrolled in Medicaid for less than 6 months, have seen a provider within 5 days of their diagnosis, and have been transferred to another hospital for treatment. CONCLUSIONS: The high prevalence of perforated appendicitis in Colorado children with Medicaid coverage is not associated with race or physical proximity to care but may be associated with the duration of Medicaid coverage, which highlights the importance of establishing medical homes to direct patients on where and how to seek care.

Training Mentor-Mentee Pairs to Build a Robust Culture for Mentorship and a Pipeline of Clinical and Translational Researchers: The Colorado Mentoring Training Program.

The Colorado Mentoring Training program (CO-Mentor) was developed at the University of Colorado Anschutz Medical Campus in 2010, supported by the Colorado Clinical and Translational Sciences Institute. CO-Mentor represents a different paradigm in mentorship training by focusing equally on the development of mentees, who are valued as essential to institutional capacity for effective mentorship. The training model is unique among Clinical and Translational Science Award sites in that it engages mentors and mentees in an established relationship. Dyads participate in 4 day-long sessions scheduled throughout the academic year. Each session features workshops that combine didactic and experiential components. The latter provide structured opportunities to develop mentorship-related skills, including self-knowledge and goal setting, communication skills (including negotiation), "managing up," and the purposeful development of a mentorship support network. Mentors and mentees in 3 recent cohorts reported significant growth in confidence with respect to all mentorship-related skills assessed using a pre-post evaluation survey (P = .001). Mentors reported the most growth in relation to networking to engage social and professional support to realize goals as well as sharing insights regarding paths to success. Mentees reported the most growth with respect to connecting with potential/future mentors, knowing characteristics to look for in current/future mentors, and managing the work environment (e.g., prioritizing work most fruitful to advancing research/career objectives). CO-Mentor represents a novel approach to enhancing mentorship capacity by investing equally in the development of salient skills among mentees and mentors and in the mentorship relationship as an essential resource for professional development, persistence, and scholarly achievement.

The demise of Oregon's Medically Needy program: effects of losing prescription drug coverage.

BACKGROUND: In January 2003, people covered by Oregon's Medically Needy program lost benefits owing to state budget shortfalls. The Medically Needy program is a federally matched optional Medicaid program. In Oregon, this program mainly provided prescription drug benefits. OBJECTIVE: To describe the Medically Needy population and determine how benefit loss affected this population's health and prescription use. DESIGN: A 49-question telephone survey instrument created by the research team and administered by a research contractor. PARTICIPANTS: A random sample of 1,269 eligible enrollees in Oregon's Medically Needy Program. Response rate was 35% with 439 individuals, ages 21-91 and 64% women, completing the survey. MEASUREMENTS: Demographics, health information, and medication use at the time of the survey obtained from the interview. Medication use during the program obtained from administrative data. RESULTS: In the 6 months after the Medically Needy program ended, 75% had skipped or stopped medications. Sixty percent of the respondents had cut back on their food budget, 47% had borrowed money, and 49% had skipped paying other bills to pay for medications. By self-report, there was no significant difference in emergency department visits, but a significant decrease in hospitalizations comparing 6 months before and after losing the program. Two-thirds of respondents rated their current health as poor or fair. CONCLUSIONS: The Medically Needy program provided coverage for a low-income, chronically ill population. Since its termination, enrollees have decreased prescription drug use and increased financial burden. As states make program changes and Medicare Part D evolves, effects on vulnerable populations must be considered.

Demystifying the NIH grant application process.

The process of applying to the National Institutes of Health (NIH) for grant funding can be daunting. The objective of this article is to help investigators successfully navigate the NIH grant application process. We focus on the practical aspects of this process, which are commonly learned through trial and error. Our target audience is generalist faculty and fellows who are applying for NIH funding to support their career development or a clinical research project.

Risk Factors of Prescription Opioid Overdose Among Colorado Medicaid Beneficiaries.

UNLABELLED: This study aims to determine risk factors of opioid overdose among the Colorado Medicaid population. A retrospective nested case-control study was undertaken. Medicaid beneficiaries who had ≥1 medical claim for an emergency department visit or a hospitalization associated with an opioid overdose from July 2009 to June 2014 were defined as cases. Controls were selected using a nearest neighbor matching without replacement. The matched controls were selected on the basis of age, sex, and opioid prescription. One case was matched with three controls. Multivariate conditional logistic regression was used to compare risk factors. A total of 816 cases with 2,448 controls were included. Six factors were associated with opioid overdose: mean morphine dose equivalent (>50 mg/d; odds ratio [OR] = 1.986 [95% confidence interval [CI], 1.509-2.614]), methadone use (switching opioid to methadone vs. no methadone use; OR = 7.230 [95% CI, 2.346-22.286]), drug/alcohol abuse (OR = 3.104 [95% CI, 2.195-4.388]), other psychiatric illness (OR = 1.730 [95% CI, 1.307-2.291]), benzodiazepine use (OR = 2.005 [95% CI, 1.516-2.652]), and the number of pharmacies used by the beneficiary (≥4 pharmacies vs. 1 pharmacy; OR = 1.514 [95% CI, 1.003-2.286]). In conclusion, several factors are associated with opioid overdose. States and communities should ensure the availability of at-home intranasal naloxone for overdose rescue on the basis of the presence of risk factors. PERSPECTIVE: This article presents the risk factors of opioid overdose among the Colorado Medicaid population. On the basis of study findings, Colorado Medicaid is currently working with physicians, hospitals, and other health system stakeholders to continue to develop policies to identify and assist this subset of our population. One such policy will be to provide at-home intranasal naloxone for overdose rescue.

Teaching systems-based practice to residents by using independent study projects.

The changing health care marketplace requires new graduates to be familiar with complex health systems. Since 1999 the Oregon Health & Science University internal medicine residency program has offered a seminar-based health systems curriculum, but residents lacked an opportunity to actively apply learned concepts. To achieve this goal, the authors developed a second curricular component, the Health Systems Independent Study Project (HSISP). The HSISP is a self-directed assessment of a health care system or delivery issue residents complete in their third year. The curriculum was implemented in 2001 with goals of gaining a focused understanding of a health systems concept and building related skills. Residents present their study projects to all curriculum participants. Topics addressed so far include adherence to coronary artery disease guidelines in a defined population, screening strategies for lung cancer, resident burnout, and many others. Residents have found that these projects enhance their systems knowledge, help them link systems-based-practice concepts to clinical practice, and foster opportunities for networking and early career development.

Importance of the medicaid medical directors' multistate collaborative for improving care in medicaid.

INTRODUCTION: There are many benefits of multistate collaboratives or networks to states, but at the center is that they allow for the opportunity to learn from other states and experts about the practices and policies states have implemented without the significant time lag of published research. This commentary examines these benefits and illustrates the importance of quality improvement collaborations to decision-making in state Medicaid programs. BACKGROUND: In 2007, the Medicaid Medical Directors Learning Network (MMDLN) began conducting quality improvement studies using their own state-level administrative data to better understand the major clinical issues facing the Medicaid populations and to work together on policies to improve outcomes. RATIONALE AND RESULTS: The three issues selected by MMDs for quality improvement monitoring to date involved an important national problem - including both morbidity and cost - and were amenable to policy solutions. The studies examined the use of antipsychotic medication in children, hospital admissions and readmissions, and early elective deliveries (i.e., elective deliveries occurring before 39 weeks). IMPORTANCE AND UTILITY: The multistate clinical quality projects conducted offer a key mechanism for achieving the goal of helping the Medicaid program deliver value-driven, high-quality, cost-effective health care in an efficient manner. These projects also provide the participating states with data to inform policies internally. CONCLUSIONS: In order for the quality of health care to improve, the system needs to be structured as a learning health care system; one that is always accessing evidence, implementing a variation of it (i.e., with new data sources or tools such as electronic clinical data), assessing effectiveness, and sharing results for others to repeat the cycle.

Medicaid admissions and readmissions: understanding the prevalence, payment, and most common diagnoses.

Reducing hospital readmissions is a way to improve care and reduce avoidable costs. However, there have been few studies of readmissions in the Medicaid population. We sought to characterize acute care hospital admissions and thirty-day readmissions in the Medicaid population through a retrospective analysis in nineteen states. We found that Medicaid readmissions were both prevalent (9.4 percent of all admissions) and costly ($77 million per state) and that they represented 12.5 percent of Medicaid payments for all hospitalizations. Five diagnostic groups appeared to drive Medicaid readmissions, accounting for 57 percent of readmissions and 49 percent of hospital payments for readmissions. The most prevalent diagnostic categories were mental and behavioral disorders and diagnoses related to pregnancy, childbirth, and their complications, which together accounted for 31.2 percent of readmissions. This analysis, conducted through the Medicaid Medical Directors Learning Network, allows Medicaid medical directors to better understand the nature and prevalence of hospital use in the Medicaid population and provides a baseline for measuring improvement.

Comparative-effectiveness research to aid population decision making by relating clinical outcomes and quality-adjusted life years.

BACKGROUND: Comparative-effectiveness research (CER) at the population level is missing standardized approaches to quantify and weigh interventions in terms of their clinical risks, benefits, and uncertainty. OBJECTIVES: We proposed an adapted CER framework for population decision making, provided example displays of the outputs, and discussed the implications for population decision makers. METHODS: Building on decision-analytical modeling but excluding cost, we proposed a 2-step approach to CER that explicitly compared interventions in terms of clinical risks and benefits and linked this evidence to the quality-adjusted life year (QALY). The first step was a traditional intervention-specific evidence synthesis of risks and benefits. The second step was a decision-analytical model to simulate intervention-specific progression of disease over an appropriate time. The output was the ability to compare and quantitatively link clinical outcomes with QALYs. CONCLUSIONS: The outputs from these CER models include clinical risks, benefits, and QALYs over flexible and relevant time horizons. This approach yields an explicit, structured, and consistent quantitative framework to weigh all relevant clinical measures. Population decision makers can use this modeling framework and QALYs to aid in their judgment of the individual and collective risks and benefits of the alternatives over time. Future research should study effective communication of these domains for stakeholders.

State Medicaid programs did not make use of prior authorization to promote safer prescribing after rosiglitazone warning.

State Medicaid programs use preferred drug lists to help limit prescribing of high-cost drugs and, at the same time, to free providers from having to obtain prior authorization for a given prescription. We examined the impact of the Food and Drug Administration's May 2007 safety warning regarding rosiglitazone (Avandia), a diabetes drug found to raise the risk of heart attacks, on the drug's availability on state Medicaid preferred drug lists and on the prescribing of diabetes medications more generally for Medicaid beneficiaries. Nearly all state Medicaid programs covered rosiglitazone as a preferred drug, requiring no prior authorization, with minimal change after the safety warning. At the same time, the safety warning was associated with a greater-than-expected decline in rosiglitazone prescribing among states providing coverage as a preferred drug. This suggests that providers reacted to the safety warning by reducing prescriptions. However, Medicaid programs that did provide coverage of rosiglitazone as a preferred drug still exhibited prescribing rates that were three to five times greater than rates in programs that did not provide coverage without prior authorization. We conclude that state Medicaid programs missed important opportunities to promote safer, more effective prescribing in the wake of the 2007 safety warning about rosiglitazone by making full use of preferred drug lists and prior authorization programs.

Improving state Medicaid policies with comparative effectiveness research: a key role for academic health centers.

After the Patient Protection and Affordable Care Act is fully implemented, Medicaid will be the largest single health care payer in the United States. Each U.S. state controls the size and scope of the medicine benefit beyond the federally mandated minimum; however, regulations that require balanced budgets and prohibit deficit spending limit each state's control. In a recessionary environment with reduced revenue, state Medicaid programs operate under a fixed or shrinking budget. Thus, the state Medicaid experience of providing high-quality care under explicit financial limits can inform Medicare and private payers of measures that control per-capita costs without adversely affecting health outcomes. The academic medicine community must play an expanded role in filling evidence gaps in order to continuously improve health policy making among U.S. states. The Drug Effectiveness Review Project and the Medicaid Evidence-based Decisions Project are two multistate Medicaid collaborations that leverage academic health center researchers' comparative effectiveness research (CER) projects to answer policy-relevant research questions. The authors of this article highlight how academic medicine can support states' health policies through CER and how CER-driven benefit-design choices can help states meet their cost and quality needs.

Exploring physician decisions about end-of-life opiate prescribing: a qualitative study.

BACKGROUND: Opiates are commonly used for symptoms at the end of life (EOL). Little is known about the decision-making process physicians go through when deciding to prescribe opiates for their EOL patients. The study's objective was to explore physician factors affecting EOL opiate prescribing. METHODS: Qualitative study of 38 physicians in the Denver area in the specialties of outpatient and inpatient medicine, geriatrics, oncology, and palliative care. Semi-structured qualitative interviews by trained interviewers asked physicians about their knowledge, attitudes, and experiences in prescribing opiates, reasons for prescribing opiates, barriers to prescribing opiates, changes in prescribing habits, and perceived patient factors that influence prescribing. Interviews were analyzed using ATLAS.ti qualitative analysis software and independently coded by two reviewers. RESULTS: We found a spectrum of beliefs ranging from the viewpoint that opiates are underused at EOL to overused. We found five key themes: practices in when and how to use opiates, barriers to prescribing, personal experiences drive prescribing, social meaning of opiates, and differences in the role of physician. Physicians interviewed described experiences, both personal and professional, that influenced their opiate-prescribing habits. All respondents expressed positive experiences with prescribing opiates in being able to ease patients' suffering at EOL and to improve their functionality and quality of life. CONCLUSIONS: Differences in prescribing habits, attitudes, and experiences of physicians influence opiate prescribing, which may lead to over- and underprescribing. Knowledge, barriers, and fears about EOL opiate prescribing need to be addressed to ensure EOL patients are receiving appropriate symptom relief.

Comparative mortality among Department of Veterans Affairs patients prescribed methadone or long-acting morphine for chronic pain.

Data on comparative safety of opioid analgesics are limited, but some reports suggest disproportionate mortality risk associated with methadone. Our objective was to compare mortality rates among patients who received prescribed methadone or long-acting morphine for pain. This is a retrospective observational cohort drawn from Department of Veterans Affairs (VA) health care databases, January 1, 2000, to December 31, 2007. We included 28,554 patients who received methadone and 79,938 who received long-acting morphine from VA pharmacies. Compared with those who received long-acting morphine, patients who received methadone were younger, less likely to have some medical comorbidities, and more likely to have psychiatric and substance use disorders. Patients were stratified into quintiles according to propensity score; the probability of receiving methadone was conditional on demographic, clinical, and VA service area variables. Overall propensity-adjusted mortality was lower for methadone than for morphine. Hazard ratios varied across propensity score quintiles; the magnitude of the between-drug difference in mortality decreased as the propensity to receive methadone increased. Mortality was significantly lower for methadone in all but the last quintile, in which there was no between-drug difference in mortality (hazard ratio=0.92, 95% confidence interval=0.74, 1.16). Multiple sensitivity analyses found either no difference in mortality between methadone and long-acting morphine or lower mortality rates among patients who received methadone. In summary, we found no evidence of excess all-cause mortality among VA patients who received methadone compared with those who received long-acting morphine. Randomized trials and prospective observational research are needed to better understand the relative safety of long-acting opioids.

Variation in pain medication use in end-of-life care.

BACKGROUND: Pain is a common and distressing symptom at the end of life that medications can help relieve. We sought to explore variation in approaches to pharmaceutical management of pain among hospice-eligible patients and to determine if variation was explained by patient or site of care characteristics. Variation in medication use may suggest areas for best practices or quality improvement in medication use in end-of-life care. METHODS: We conducted a secondary analysis of randomized trial data, examining use of five medication classes: opiates, nonsteroidal anti-inflammatory drugs (NSAIDs), adjuvant pain medications (tricyclics and antiseizure), stimulants, and antianxiety medications in 16 study sites nationwide. Descriptive statistics were generated for patient-level data and by site. Unadjusted and adjusted odds ratios were calculated to compare patient and location of care characteristics with each medication class use by site. RESULTS: We found variation in medication use was not predicted by most patient characteristics or location of care (home versus facility). Use of all types of pain medications decreased with age (odds ratio [OR] 0.75 [0.63-0.90]). Medication use varied between sites: a range of 14%-83% of patients were on different types of opiates, 0%-40% on NSAIDS, 20%-69% on benzodiazepines, 0%-25% on adjuvant medications, and 0%-23% were on acetaminophen at any time during the data collection period. CONCLUSIONS: Pain and adjuvant medication use differs widely by site of care. Further research is needed to determine the extent to which provider and patient choice contribute to prescribing variation, and to explore associations between patient symptoms, medication variation, and patient care quality.

Influencing public policy to improve the lives of older Americans.

Aging of the U.S. population raises numerous public policy issues about which gerontological researchers, policy experts, and practitioners have much to contribute. However, the means by which aging-related public policy is influenced are not always apparent. Drawing on experience working in the U.S. Senate and other settings as Health and Aging Policy Fellows, the authors outline the formal and informal processes by which public policy is shaped in the U.S. Congress. Many who seek to influence public policy do so by telling legislators what they want. A less obvious path to policy influence is for gerontologists to offer their expertise to legislators and their staff. The authors provide specific recommendations for how gerontologists can establish productive and ongoing relationships with key legislative players. The authors also emphasize the importance of collaboration with advocacy groups and with local and state stakeholders to advance aging-related public policy to improve the lives of older Americans.