Causes and Consequences
of Health Care Efficiency

Mission of the Grant

Our original PO1 grant proposal was motivated by the remarkable differences in health care intensity in the U.S. The goal is to understand and improve efficiency, the target is a more actionable level fro physicians and the hospitals where they work, and the means include collaborations with key stakeholders to design successful interventions. The projects are:

1. Measuring the Efficiency of Health Care Providers. This project develops measures of efficiency – both quality and costs – and explores their implications at the level of primary care physicians, physician groups, and hospitals. We use Medicare data, patient and physician surveys, and specialty board certification scores from the American Board of Internal Medicine and the American Board of Family Practice.
2. Technological Growth and the Efficiency of Health Care Diffusion. We study the contribution of “high-tech” health care to cost growth, the diffusion of different types of medical innovations, whether high-quality health providers affect population-level health outcomes, and how provider networks affect diffusion.
3. Geography and Disparities in Health and Health Care. We propose hospital-level disparity “report cards,” to study sorting of patients to disproportionately Black hospitals and to test whether the preferences of patients from a recent nation-wide survey of Medicare enrollees are reflected in their treatments. Finally, we estimate how racial differences in surgical rates affect disparities in actual health outcomes.

Cores: The Cores include an initiative to work with seven leading hospitals to address inefficiency in the provision of care, data support, and health efficiency measures for users of the Health and Retirement Study.

Relevance: The project seeks to understand why efficiency is so poor in the U.S. health care system: 30% of health care spending is wasted, and quality of care is seriously lacking. We also explore several approaches to improving quality and reducing unwarranted variations in expenditures.

Our current Plans

1. Developing new Medicare claims datasets. With our data analysts, we have made important strides in updating cohorts of datasets in line with each of the projects, in particular the heart attack, colon cancer, and hip fracture cohorts, which have been defined now through 2005. These also include measures of physician “groups” that can be used in developing pay-for-efficiency policies.
2. Patient survey. We are planning, in conjunction with the Center for Survey Research, a patient survey focused on people who have recently had operations for breast cancer, stable angina (e.g., stents), and prostate cancer to determine how well they understood the risks and benefits of their surgery. We expect to go into the field – assuming CMS allows us to contact patients – this year.
3. ABIM and ABFM Data. We continue to work with the American Board of Internal Medicine and American Board of Family Medicine to use their data on individual physician board scores to further understand the role of physician training and competence in affecting health care outcomes and costs.
4. Measures of Health Care Intensity. We are actively seeking to improve our cross-regional measures of health care spending through a new program of price adjustments. This will allow us to compare “pure” utilization differences between (say) Phoenix and Philadelphia – despite the very different rates at which Medicare reimburses in these two regions.
5. Further study of growth rates in health care expenditures. A primary goal is to better understand what factors are causing growth in health care costs. These require a complete merging of the longitudinal data and allocation into specific “bins” to facilitate this accounting exercise.