Skip to Main Content

Link Between Health Care Spending and Health Outcomes Elusive

November 07, 2019

You might think hospitals that spend the most on treatment achieve the best outcomes for their patients.

But that isn’t always the case.

Jonathan Skinner, the James O. Freedman Presidential Professor of Economics at Dartmouth College and director of the Aging Program at the National Bureau of Economic Research, told a gathering in Winslow Auditorium on Wednesday that the link between health care spending and health outcomes is more elusive than it may appear.

Treatment decisions, hospital resources, the number of doctors involved in a patient’s care and even the hospital’s location are all factors influencing costs and health outcomes, said Skinner, who delivered this year’s Dean’s Lecture for the Yale School of Public Health’s Program on Aging Seminar on Nov. 6.

Rather than mapping a neat arc that shows hospital health outcomes improving in direct relation to spending on care, researchers like Skinner have found the actual results are all over the place. Some hospitals with lower costs have exceptional patient outcomes while others, with much higher costs, achieve poorer results.

The key question, Skinner said, is why such differences exist? The short answer—“production inefficiency.”

In an as yet unpublished study, Skinner compared treatment costs and health outcomes at various hospitals that treated 500 or more Medicare patients for heart attacks or acute myocardial infarction (AMI).

Skinner found that hospitals with lower treatment costs and higher survival rates for AMI patients—basically the “highly efficient” hospitals—employed a greater use of effective treatments. That group included teaching hospitals and hospitals that require statin medication after discharge, visits with a primary care provider within two weeks of the event and same-day balloon angioplasty to open clogged arteries.

Hospitals with higher treatment costs and less favorable outcomes for AMI patients—the “less efficient” ones—utilized less-effective treatments. Those practices included expensive extended home health care post discharge, assigning multiple physicians and specialists to a patient and using double CT-scans with and without contrast to check a patient’s condition. Skinner said the latter practice is an outdated procedure that “costs a lot and provides no additional health benefit.”

Skinner’s hour-long lecture before a packed house of faculty and students provided a clear illustration of the value of health economics and the study of health policy and management, which is one of the school’s core academic and research areas.

Skinner found that even location can influence health care costs and outcomes. Two of the higher cost, lower performing hospitals in his study, Hialeah Hospital and North Shore Medical Center, are both located in Miami, where home health care is very popular and very expensive, he said.

To be fair, Skinner said his study only compared costs and outcomes. He did not dig deeper into all of the possible factors contributing to the disparities. And the findings only relate to treatment for AMI. The study does not reflect hospitals that may provide low-cost, high quality treatment in one area but do poorly in another. Skinner compared his findings to the classic conundrum of Schrödinger’s cat.

“Some hospitals may be doing really well in one area and bad in another at the same time,” said Skinner.

The bottom line, Skinner said, is how a hospital is managed.

"There is different internal productivity within hospitals,” Skinner said. “Some just do better with what they’ve got.”

Submitted by Colin Poitras on November 07, 2019