Demonstrating Success: How to Improve Care and Pay Physicians More While Reducing Costs

The cost of healthcare in the United States is staggering. This is well-documented but bears repetition. In 2012, healthcare in the United State cost $2.8 trillion. Of this, 75% was related to chronic conditions.

That 75% amounts to $2.1 trillion spent annually on treating chronic conditions.

Given this cost, we have speculated before that improvements in chronic conditions could be achieved through reducing administrative spending. This reasoning was based on an intuition that the U.S. healthcare system is overburdened with administrative costs (we lead the world in this department), has too few doctors (30% less than OECD average), and has the highest per capita health spending (over 1.5 times that of the second-closest nation).

Intuition may be a great starting point but in the complicated, expensive realm of healthcare, greater evidence is required.

Fortunately, the Centers for Medicaid and Medicare Services (CMS) initiated a demonstration project in 2005 that focused on prevention and chronic disease management. The project ran from 2005 to 2010 and showed that improving quality and lowering costs is indeed possible. Administrative costs were not considered as part of the demonstration, although the project does support the notion that they are rife with opportunity for improvement.

The demonstration project was collaboration between CMS and ten physician groups. The goal of the project was to achieve demonstrable savings in four quality areas: diabetes, congestive heart failure, coronary artery disease, and preventive care. Any savings realized by CMS were shared 80/20 with the physician groups.

The results?

By year five, every physician group met 30 of the 32 quality improvement measures. Four physician groups generated over $36 million in savings (with $29 million in incentive payments). Previous years showed similar success.

For at least those four physician groups, it seems the intuition is born out. More detail is needed along with a more careful examination of results, but this is a promising start.
Of more concern than economic savings is the care patients received. Here too, the metrics demonstrate improvement across the board, averaging a near 10% improvement in quality metrics. CMS reports:

The PGPs have increased their quality scores from baseline to performance year 5 an average of 11 percentage points on the diabetes measures, 12 percentage points on the heart failure measures, 6 percentage points on the coronary artery disease measures, 9 percentage points on the cancer screening measures, and 4 percentage points on the hypertension measures.

It is possible, then, to improve quality while achieving savings. And, as noted, there is room for additional improvement by examining ways to reduce and eliminate the administrative overhead involved in managing this process.

While improvements around 10% are indeed praiseworthy, there is a sobering reality here. Martin Gaynor’s blog entry on pay for performance points to a key problem. In one study referenced here, Nyweide et al. found that “Relatively few primary care physician practices are large enough to reliably measure 10% relative differences in common measures of quality and cost performance among fee-for-service Medicare patients.”

For smaller practices and even individual physicians, the improvements achieved may not even be measurable. That is significant. As we have previously observed, 94% of healthcare professionals would not recommend entering the field. With dissatisfaction this widespread, long-term sustainable improvements are only possible with clinician buy-in, which necessitates discernible results.

One aspect that would be discernible is the incentive payment. The $29 million in payments represents about $13,000 per physician at the four physician groups. (This is an admittedly facile calculation that uses a simple proportion of the 5,500 physicians in the groups and neglects to account for other clinical personnel.)

This results in per patient savings amounts of about $400. Given per capita healthcare spending of about $8,800 per year, this savings represents 4.5% of average cost.


This chart from 2009 shows the distribution by population of healthcare spending. This chart represents personal healthcare spending, but the distribution both here and for total healthcare spending has remained remarkably consistent through the years. As the chart makes clear, healthcare spending is heavily skewed toward very few high-cost patients. 1% of the population accounts for 20% of healthcare costs; 5% of population accounts for 50% of healthcare costs; and 50% of population accounts for 97% of healthcare costs. Given these metrics, it seems like a 10% improvement in chronic disease quality measures should result in more than a 4.5% reduction in cost.

From a purely economics point of view, the goal of preventive medicine and improved maintenance care is to avoid, defer, or reduce costly procedures and complications that drive up the right end of the curve. From a human point of view, the goal is to improve the quality of life while avoiding potentially needless pain and complications for patients.

The lesson to be gained from the demonstration project is that if ensuring the best care possible for these chronic conditions results in only a 5% decrease in healthcare costs, there is much more to improve upon than simply physician efficiency and patient follow-up. In order to truly tackle an overall reduction of healthcare spending, we need to simplify the larger processes and make the entire system more efficient by focusing on reducing administrative costs at all levels of the system.