Taming Data for Useful Improvements

Quality metrics are becoming more important for healthcare providers. Accountable care, Meaningful Use, physician quality reporting and other programs are driven by quality measures.

For more than 20 years, the Dartmouth Atlas Project has documented glaring variations in how medical resources are distributed and used in the United States. The project uses Medicare claims data to provide comprehensive information and analysis about different markets, as well as individual hospitals and their affiliated physicians.

David Goodman, MD, co-principal investigator of the organization, spoke with Clinical Innovation + Technology about the growing importance of quality metrics and how stakeholders can use Dartmouth Atlas data to improve their care delivery.

CIT: How has healthcare data collection changed since the Dartmouth Atlas Project first began?
DG: There is tremendous interest in public reporting on quality data today. When the Dartmouth Atlas Project first got started, it was the only source of data that spanned the nation while providing local quality metrics. Now, there are many different sources including the Centers for Medicare & Medicaid Services (CMS), which really has done a great job in providing metrics.

How do different stakeholders use the Dartmouth Atlas data?
Atlas data, along with other public data sources, provide providers with a view of their care that they really can’t see at the ground level. It’s very difficult to tell the quality of care that I deliver. We all do our best every day, but we need external sources of data that help us check how we’re doing to either benchmark locally or to the rest of the country. Our data make visible what is invisible to doctors and nurses.

It’s hard to measure patient use of these data. Certainly, some of our data are very relevant to patients in terms of showing the quality of care by region or by hospital. Some measures are specific to certain types of care, such as end-of-life care for cancer patients where there is market variation in the degree of aggressiveness or the amount of palliative care patients receive.

There also are interesting data about places that have very low or very high procedure rates. We know that these extreme variations in procedures rates are largely driven not by patient health status preferences but by physician practices in a given health system. That serves as a stimulus for patients to ask questions about the full range of options available for their care.

How will your data capture plans change in the future?
Dartmouth Atlas data relies largely, but not entirely, on Medicare claims data. We are beginning projects for the under 65 population.

There are other data sources, such as the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) surveys from CMS in which hospitals are strongly incentivized to query patients about their hospital experiences such as whether the clinicians were communicating well, whether their pain was well controlled and whether their room was clean. CMS also has quality measures specific to hospitals. Although hospitals are not absolutely required to provide the data, the incentives to do so are very, very high. Most patients are now providing quality data about certain types of care, such as pneumonia and heart attack, that we sometimes use. It’s well worth the read.

How will the ever-growing data from EHRs impact your work?
The direction of both measures of quality and utilization outcomes are going to expand in the use of large administrative datasets in the under 65 population like data we have from Medicare. Data from patient-recorded outcomes measures are being developed and slowly being implemented. We’ll begin to see some pilot projects that draw upon data from EHRs. They are the most important data because they report meaningful patient experiences. The expectation is that these data will be increasingly available for research and for public reporting.

What are Dartmouth Atlas’s top priorities going forward?
Our top priorities are to continue aggressively into the under 65 population. We are doing that with projects that examine orthopedic care, overall cost of care and pediatric care. Also, on the Medicare side, we have initiated a large project looking at patterns of drug prescribing for Medicare beneficiaries.

Dartmouth Atlas work has always been linked to remedies for improvement. Our institute is actively engaged in the development of clinical microsystem improvements, improving patient decision making through informed patience choice and in redesigning ideas for the care delivery system. It is this research and epidemiological data that stimulated ideas for how we might improve care, particularly Elliott Fisher and his accountable care learning collaborative and other efforts he leads at the Dartmouth Institute.