 |
||
 THE GRANTThe goal of this project is to develop methods to track disparities in the effective delivery of interventions throughout the state of Washington in order to improve population health. The National Institute on Aging (NIA) Principal Investigators:Tracking Disparities in the Effective Delivery of Health ServicesHealth systems improve the health of populations in part by delivering high-quality interventions to those who need them. Policymakers, managers, and providers of health services at all levels frequently demand to know how well they are accomplishing this task and to what effect. Two different metrics integrate aspects of intervention delivery to allow comparisons across interventions, individuals, and populations – coverage and effective coverage. Frequently in the US, health services are provided by, and the responsibility of, local government. In many instances, this is the county government. Counties and the populations within them differ markedly from one another, as do the counties' health systems performance. The goal of this project is to develop, apply, and refine methods that enhance the ability to track disparities in the effective delivery of interventions to improve population health. To do so, IHME needs a test area in which to work. In this project, we will use all available data to estimate the coverage and effective coverage of a range of key health interventions for each of the 39 counties in the state of Washington and examine trends from 2002-2007. These 39 counties represent a diverse array of subpopulations and differ in terms of overall population, racial/ethnic makeup, and socioeconomic status. Interventions to be studied range from inpatient management of major diseases – including myocardial infarction and stroke – to outpatient management of risk factors such as using lipids to control high blood pressure or screening for major cancers. The state of Washington is our test case for several reasons, including the fact that the state has a robust reporting system for data collection at the hospital level and has several centrally orchestrated data sources that complement federal data collection efforts. This provides us a data-rich environment in which to perform analyses and refine techniques to achieve the most reliable and high-caliber results. From this test environment, we aim to develop a set of standardized tools and protocols that can be used as a template for estimating effective coverage in other settings. These tools can then be applied to estimate effective coverage of key interventions in all US counties and in international settings to estimate effective coverage at the subnational level. Defining CoverageIn the US, the term coverage often refers to insurance coverage – the fraction of the population that has health insurance. Intervention coverage captures the utilization of a given intervention by the population in need. Effective coverage adds a third critical component – quality, defined by the fraction of the optimum health gain realized by those receiving the intervention. For example, if 80 out of 100 children receive measles vaccinations, but only 90% of those 80 children are effectively protected against measles, then the intervention coverage would be 80%, but effective coverage would be 72% (90% quality multiplied by 80% intervention coverage). This same logic can be extrapolated to the health system as a whole, taking into account a range of interventions aimed at addressing prevalent health needs across the population, and used to calculate a comprehensive, system-level effective coverage assessment. Relevant Technical PointsFrom a technical standpoint, we will identify three population-based components:
We will develop strategies to estimate the need for an intervention and then correct for diagnosis and bias in self-reporting. We will use a combination of data sources to estimate utilization – including self-reports in surveys, drug inventories, and health service registries. We will estimate quality in several ways. For one set of interventions, we will analyze survey questions from the National Health and Nutrition Examination Survey and the Washington State Cardiovascular Health Examination Survey, among others, which demonstrate changes in measured biological indicators, such as blood pressure. We also will assess self-reported and measured data on functional health status, such as a change in vision with the use of visual aids. We will combine estimates of coverage and effective coverage from multiple data sources, including administrative records that include measured and self-reported data. Finally, we will use small area estimation methods to produce estimates of coverage and effective coverage at the county level. CollaboratorsIn our effort to best understand the differences among counties and to include all available data to produce the best possible estimates, we will convene interested representatives from the county health organizations. We hope to learn from their experience and expertise and will share and discuss our results with them as the project unfolds. We also will reach out to a group of interested parties from other states and countries once we have a working model and preliminary results to discuss ways in which the approach could be extrapolated to other settings. For more information, please contact: us@healthmetricsandevaluation.org This project is supported by award number RC1AG035616 from the National Institute on Aging, which is part of the US National Institutes of Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute on Aging or the National Institutes of Health. |
