Economic Analysis and Quality Improvement in Public Health

The adoption and use of quality improvement (QI) techniques has grown rapidly among agencies that deliver public health programs and services in recent years. Recession-driven reductions in governmental spending has fueled this movement, spurring agencies to seek ways of ‘doing more with less’ to preserve their operations. The newly-launched national accreditation program for public health agencies (PHAB) has also encouraged the trend by creating accreditation standards specifically tied to QI adoption and implementation. And the federal Affordable Care Act has created strong incentives for QI in public health, most notably through the CDC’s National Public Health Improvement Initiative (NPHII), which provides direct federal funding and technical assistance to state and local public health agencies to support QI applications.

What’s the impact of this flurry of activity? To be sure, scientific evidence concerning the effectiveness of QI in producing health and economic benefits is not clear cut within the medical care sector. Hospitals and health systems have been using QI techniques for several decades now, including statistical process control analyses, quality councils, “lean” production methods, PDSA cycles, and collaborative models of improvement such as those championed by the influential Institute for Healthcare Improvement led by former CMS administrator Don Berwick. While there are certainly individual examples of QI success within specific health care organizations, overall improvements in quality, efficiency, and equity in medical care have been modest at best over this time frame. After a long period of waiting for system-wide impact from voluntary QI efforts coordinated through entities like the national network of federally-funded quality improvement organizations (QIOs), the federal government has begun to move beyond QI to implement much stronger policy mechanisms like public reporting, pay-for-performance, and accountable care organizations.

A recent analysis by Harvard University’s Robert Kaplan and Michael Porter points to one reason for the lackluster performance of QI in medical care delivery: the lack of detailed and high-quality data on costs. The authors note in their Harvard Business Review paper:

“Poor costing systems have disastrous consequences. It is a well-known management axiom that what is not measured cannot be managed or improved. Since providers misunderstand their costs, they are unable to link cost to process improvements or outcomes, preventing them from making good decisions….Poor cost measurement [leads] to huge cross-subsidies across services…Finally, poor measurement of costs and outcomes also means that effective and efficient providers go unrewarded.”

Without good cost information, health care organizations are vulnerable to focusing their QI time and attention on the wrong problems, and to reaching incorrect conclusions about which solutions work best. Now public health agencies are joining the QI movement in droves, but unfortunately, data on the costs of delivering public health programs and services are even more limited than data on medical care costs. Is public health QI destined to experience the same slow progress and lack of system-wide impact as in medical care?

The good news is that some important initiatives are now underway to fill this void in research-quality knowledge about the costs of public health delivery – initiatives that hold great promise for advancing the goals of QI. I had the opportunity to speak on a panel examining of this progress at this week’s Open Forum on Quality Improvement in Public Health held in Memphis, funded by the Robert Wood Johnson Foundation and organized by the National Network of Public Health Institutes. CUNY’s Marthe Gold opened the panel by summarizing the recommendations of an influential Institute of Medicine report on public health funding that has guided many of the efforts undertaken over the past year to improve our understanding of the economics of public health delivery. Marthe noted that among the key recommendations from this report were to (1) develop a national chart of accounts for public health that will enhance the measurement of how funds are used within the governmental public health system at federal, state, and local levels; (2) expand research on the costs, cost-effectiveness, and value of public health programs and services; and (3) identify the components and costs of a minimum package of public health programs, services, and capabilities that are recommended to be available in every U.S. state and community.

My talk (slides here) on this panel focused on some the approaches now being used for cost estimation in public health delivery in response to the IOM recommendations. Some of these approaches are now being tested through the Delivery and Cost Studies (DACS) underway through our Public Health PBRN program, which I have posted about recently. Additionally, three states have blazed ahead with projects to estimate the costs of a “minimum package” or “core set” of public health programs and capabilities on a statewide basis. Each using a different set of methods, data and assumptions, these three projects – underway in Washington, Ohio, and Colorado – offer valuable opportunities for identifying the strengths and limitations of alternative empirical approaches to cost estimation in public health. Very much in keeping with the paradigm of states as laboratories for fiscal policy, these projects are already informing the methodologies currently under development for national-level cost estimation as recommended by the IOM.

The third speaker on our panel, Jason Orcena who directs the Union County Health Department in central Ohio (and is writing PhD dissertation at UIC on these topics), brought down the house with his description of the political economy processes and intergovernmental dynamics that shaped Ohio’s effort to reach consensus on a “minimum package” of public health services over the past year. The politics of Affordable Care Act implementation and Medicaid expansion in this political swing state were just a few of the dynamics at play during this process. His observations underscored the importance of generating credible estimates of costs and benefits attributable to public health programs and services to inform policy decision-making in politically heterogeneous environments.

The Open Forum meeting also featured examples of the progress being made in incorporating economic evaluation principles into the evaluation of QI initiatives implemented in public health settings. In particular, I was very heartened to see that growing numbers of state and local agencies using the Public Health Return on Investment Template prototype that our research center developed with ASTHO last year to facilitate economic analysis of QI initiatives. ASTHO’s Karl Ensign gave an update on the tool and how it is being used in a variety of public health settings and applications. Of course, this growing experience with using the prototype tool is uncovering many features that need expansion and improvement, so there is much more work to be done on this front. But clearly there is growing enthusiasm and activity behind the concept of integrating economic analysis and QI within the public health system.

This QI meeting marked my second trip to Memphis in as many weeks (my earlier post on the first trip), and I again left the city feeling optimistic about the progress in improving the science and practice of public health. Following the structure of any Delta blues song, there’s much to be worried about in public health, but there’s also a slow and steady movement forward.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s