David Abstract
Emergency Medical Services (EMS) networks are at the heart of our public health system. Their primary role is to respond to, stabilize, and transport patients involved in emergencies, such as myocardial infarctions, motor vehicle crashes or incidents of violence. While more than $5 billion per year are spent on EMS transport in the U.S., very little is known about the effectiveness of prehospital interventions, and even less about the role and determinants of its organization.
This grant proposal spans a number of cross‐disciplinary research projects intended to enhance our understanding of the role, structure, efficiency, and efficacy of EMS. Specifically, we are interested in four broad topics: the benefits of human capital accumulation among emergency medical technicians; the evaluation of prehospital clinical practice and its effect on patient outcomes; the determinants and consequences of EMS privatization; and the determinants of child safety seat use and its impact on injuries and EMS resources and intensity. The studies we propose here will take advantage of a unique dataset. The Mississippi Emergency Medical Services Information System (MEMSIS) records all prehospital calls for trauma between 1991 and 2005, providing detailed information on the nature and location of the incident, patient demographics and injury characteristics, time‐stamps for all prehospital stages, and the basic and advanced life support procedures performed on‐scene by emergency responders.
The research team is composed of clinicians, economists, and health services researchers. The project uses a novel and comprehensive dataset on approximately four million EMS incidents over a decade and a half. All studies will exploit recent advances in econometrics and introduce a number of improvements over the existing literature.
Draine Abstract
This proposal explores the service usage patterns for people with serious mental illness leaving jail. Over the past 20 years, including 17 years of my work, research has focused on understanding the significant increases in the numbers of people incarcerated in prison and jails, especially the numbers associated with people with mental illness. This attention has led to the investment of millions of dollars in services that are designed to link people with mental illness to existing treatment resources post release. Yet, questions remain about the mental health system’s capacity to provide comprehensive treatment services to more then a small minority of these returning prisoners. These questions are exacerbated by the paucity of information in the research literature about the type of services that are available to this client population at release, or the timing, type, intensity, and duration of received services. This shortcoming takes on special significance when the complex co-occurring problems such as substance use and homelessness that are generally associated with this client population are considered. These issues leave individuals especially vulnerable to exclusion from services in the mental health system. We propose to examine one mental health system’s capacity to provide a comprehensive array of treatment services to people with mental illness leaving jail by mapping the potential service pathways against the actual service usage patterns for people with mental illness leaving jail in order to identify gaps in available services and any incongruence between the expected and actual usage patterns.
Dahodwala Abstract
Background: Parkinson’s disease (PD) is a common cause of aging-related movement changes; however, more than 40% of affected individuals are not diagnosed and treated. Attitudes about aging and related help-seeking may affect the timely diagnosis of PD.
Objective: The aim of this study is to develop a standardized instrument with established psychometric properties to measure elderly persons’ parkinsonian symptoms, expectations about aging-related movement changes, and healthcare seeking beliefs for parkinsonian symptoms.
Methods: We will first engage experts and review the literature to generate items for a scale (content validity). Next focus groups composed of older adults will be conducted to refine the content of the scale and test face validity. Finally, community members will be surveyed to test the internal consistency, construct validity, and criterion validity of the instrument. Preliminary hypothesis testing regarding the associations of demographic, socio-cultural and health factors with expectations regarding aging and healthcare seeking for parkinsonian symptoms will be performed.
Significance: Little is known regarding why so many people with PD remain undiagnosed. If untreated, this disease dramatically decreases quality of life. Increasing our understanding of factors associated with under-diagnosis will lead to the development of interventions to improve appropriate and timely diagnoses. Specifically, depending on the results, findings from the proposed study will be used to increase physician and community awareness and change primary care screening practices. We will also be able to identify subgroups with movement problems that are at increased risk for under-identification.
Doshi Abstract
Medicare Part B has traditionally covered a limited number of outpatient prescription drugs and biologicals for all Medicare beneficiaries. A majority of these Medicare Part B covered drugs include physician dispensed drugs such as injectables. The Medicare Prescription Drug, Improvement, and Modernization Act (MMA) of 2003 significantly changed the coverage and reimbursement of outpatient drugs. First, the MMA established a new methodology for physician reimbursement of most Medicare Part B covered drugs. Second, it established Medicare Part D, which started providing comprehensive outpatient prescription drug coverage for beneficiaries from January 1, 2006. Under this new benefit, many drugs previously not covered by Part B are now being reimbursed. This pilot study will be the first to examine the impact of the changes in physician reimbursement for Part B covered drugs over time (2003-2006) and availability of outpatient drugs through Part D (2006) on Medicare Part B drug utilization and spending across 2002 to 2006. In addition, we will estimate the extent of substitution of Part B drugs with Part D covered alternatives in a cohort of patients using Part B drugs in 2005 who voluntarily enrolled into private Part D plans in 2006.
Carr Abstract
Emergency services in the United States are widely accessible through the nation’s network of emergency departments (EDs). After initial stabilization and diagnostics, however, some patients require transfer to advanced services not available locally. Substantial delays are often associated with coordinating transfer to a receiving hospital with the capability to treat the patient. Too often, opportunities to intervene are lost because of systems inefficiencies.
In a recent review of the emergency care infrastructure in the US, the Institute of Medicine (IOM) recommended the “…develop[ment] of an evidence-based categorization system for EMS, EDs, and trauma centers based on adult and pediatric service capabilities.” The express goal of this system is to facilitate an increasingly regionalized approach to managing complex emergency medical and surgical diseases. The IOM describes the objective of regionalization as “…improving patient outcomes by directing patients to facilities with optimal capabilities for any given type of illness or injury.” In part, regionalization has its roots in the health services research description of the volume-outcome relationship. In addition to patient volume, labor and resource availability are important contributors to high quality outcomes.
A regionalized approach to the management of trauma patients has been developed, implemented, and applauded by the IOM. Similar systems for the management of other time and resource intensive diseases such as acute myocardial infarction, acute ischemic stroke, sepsis, and/or post-cardiac arrest care are less developed. The US infrastructure of EDs would serve as an essential linchpin in a regionalized emergency care delivery system for these nontraumatic diseases. EDs need to be categorized according to their capabilities and the capabilities of their parent hospitals. The distinction between ED capability and hospital capability are distinct but related concepts. An ED may be well staffed 24/7 with high quality providers, but if no interventional cardiology group is reliably available for off-hours cardiac catheterization, patients in need of these services need to be transferred to a regional center. Efficient inter-hospital transport systems need to be developed to meet time benchmarks corresponding with improved outcomes.
The knowledge and attitudes of emergency physicians with respect to developing this categorization system for EDs as the centerpiece of a regionalized emergency care delivery system is unknown. We therefore plan to develop and pilot test a questionnaire of ED directors to inform this question, and ultimately lead to a much larger survey that would be representative of all ED directors in the US.
Kahn Abstract
Variation in the quality of intensive care across hospitals has prompted calls for regionalization of car for patients with critical illness. Conceptual support for regionalization comes from several recent studies showing improved outcomes in hospitals with a high volume of critical care admissions. Although regionalization has the potential to improve survival in the ICU, more information s needed before such a system can be implemented. This project will use Pennsylvania state discharge data to study the structure, feasibility and potential mortality reduction under a regionalized system of critical care. First, we will examine the organizational characteristics of low and high volume hospitals in the state and compare the demographics of patients receiving care in hospitals of different volumes. Second, we will create a preliminary geographic model of regionalization, analyzing transfer times and changes to ICU census at destination and referral hospitals. Finally we will estimate the survival benefit of inter-hospital ICU transfer under regionalization using propensity scores and instrumental variable analysis. This research will inform health policy decisions about regionalization, offer proof-of-concept for the use of instrumental variables in ICU observational studies, and provide preliminary data for a future demonstration project organization in Pennsylvania hospitals.
Hillier Abstract
We propose to form a working group composed of faculty and doctoral students from the School of Design, School of Nursing, Graduate School of Education, School of Medicine, and School of Arts and Sciences to investigate racial disparities within health care and healthcare professions training. Our work will compare conditions in Philadelphia at the turn of the 20th and 21st centuries, bringing together researchers who have been studying overlapping issues but have not previously worked together. Research on healthcare professions training and workforce development and disparities in health care is currently underway within several different schools at Penn. This project will also capitalize on historical research about W.E.B. Du Bois’ 1896 study, The Philadelphia Negro within the School of Design, historical research on the blacks and health care in the late 19th and early 20th centuries in the School of Nursing, and research about the more recent history of health care as it relates to race. Over the next 18 months, this collaboration will result in new academic research, research opportunities for undergraduate, graduate, and doctoral students, cross‐disciplinary relationships among Penn schools and researchers, teaching material for high school and college classes, and a proposal for further funding.
Polsky Abstract
There remains a significant need for the government, third-party payers, plans and insurers to accurately forecast prescription drug expenditures as well as identify future high-cost cases. Few studies have been done to study the predictability of drug spending. Previous studies have only used ordinary least squares (OLS) methods to predict prescription drug expenditures and these models have been limited in their predictive performance. Artificial neural networks (ANNs) are new mathematical models that imitate observed properties of biological nervous systems. This methodology can deal with complex systems by using non-linear flexible function forms and recognizing patterns in the dataset. While ANNs have not been previously tested for this purpose, they hold great promise for improving the prediction of drug expenditures given their successful application in other fields. Using the Medicare Current Beneficiary Survey (MCBS) Cost and Use Files, we will compare the performance of various econometric models and ANNs in forecasting prescription drug expenditures. Our study will be the first application of this cutting-edge technology in forecasting drug expenditures and has the potential to open a new area for future research regarding application of ANN models in predicting medical and prescription drug expenditures.
Robinson Abstract
Clinical quality improvement continues to be a national focus. The push toward quality has also led to a desire for transparency including public reporting and pay-for-performance. There is concern about the unintended consequences of pay-for-performance for health care providers as well as the effect the incentive system might have on health care disparities. It is unclear what effect, if any, a pay-for-performance system will have on vulnerable populations. The purpose of this study is to show that people from vulnerable populations are cared for by hospitals that perform poorly on current clinical quality indicators. This study also aims to show information technology utilization as an important factor in the ability of a hospital to achieve higher clinical quality indicator scores. The data gathered from this study will then be used as a starting point to evaluate longitudinal changes in resources and measures of quality for hospitals that care for vulnerable populations. Specifically, the first part of the study will explore the clinical quality indicator scores for hospitals that treat patients from vulnerable populations and look at the association between patients from a specific demographic population and the clinical quality indicator scores of the hospitals that serve them. The second part of the study will look at the impact of technology on clinical quality indicator scores by exploring the association between technology utilization and clinical quality indicator scores for all hospitals and then for those hospitals that care for vulnerable populations.
Pauly Abstract
The technology adoption literature often does not focus on the potential role of physicians, despite the fact that almost all medical technologies involve physician use or prescription. Assuch, there is a gap in understanding the role of physician selection in the adoption and diffusion of technology in health care. How does the uncertainty discouraging adoption balance out against the incentive for technological preeminence encouraging it? The goal of this project is to address this question by studying the interactions between surgeons’ reputation, expertise, and frequency of use and their choice of timing for adoption of a new surgical technology. We hypothesize that greater reputation and expertise with the old technology will lead the surgeon to delay adoption of an innovation that substitutes for the old technology, whereas reputation and expertise will lead to earlier adoption of new technologies that are complementary to or incrementally improve the old technology. Our conceptual framework will build from the physician’s adoption decision, to the manufacturer’s response, and result in an equilibrium of simultaneous manufacturer and surgeon action. The empirical examination will test the implications of this equilibrium. The analysis will utilize discharge data from Florida linked to detailed data on physician characteristics. A key part of the analysis will be constructing an index of reputation. Understanding technology adoption by physicians has important implications on welfare, such as explaining which innovations are realized in the market, regulatory policy toward timing of new product introductions, and physician agency issues in maximizing positive patient outcomes.
|
