In The Wall Street Journal, Scott Gottlieb of the American Enterprise Institute rounds up a batch of studies showing that health outcomes for patients covered by Medicaid, the joint federal-state health program for low-income and disabled individuals, are actually worse than those covered by no insurance at all.

• Head and neck cancer: A 2010 study of 1,231 patients with cancer of the throat, published in the medical journal Cancer, found that Medicaid patients and people lacking any health insurance were both 50% more likely to die when compared with privately insured patients—even after adjusting for factors that influence cancer outcomes. Medicaid patients were 80% more likely than those with private insurance to have tumors that spread to at least one lymph node. Recent studies show similar outcomes for breast and colon cancer.

• Major surgical procedures: A 2010 study of 893,658 major surgical operations performed between 2003 to 2007, published in the Annals of Surgery, found that being on Medicaid was associated with the longest length of stay, the most total hospital costs, and the highest risk of death. Medicaid patients were almost twice as likely to die in the hospital than those with private insurance. By comparison, uninsured patients were about 25% less likely than those with Medicaid to have an "in-hospital death." Another recent study found similar outcomes for Medicaid patients undergoing trauma surgery.

• Poor outcomes after heart procedures: A 2011 study of 13,573 patients, published in the American Journal of Cardiology, found that people with Medicaid who underwent coronary angioplasty (a procedure to open clogged heart arteries) were 59% more likely to have "major adverse cardiac events," such as strokes and heart attacks, compared with privately insured patients. Medicaid patients were also more than twice as likely to have a major, subsequent heart attack after angioplasty as were patients who didn't have any health insurance at all.

• Lung transplants: A 2011 study of 11,385 patients undergoing lung transplants for pulmonary diseases, published in the Journal of Heart and Lung Transplantation, found that Medicaid patients were 8.1% less likely to survive 10 years after the surgery than their privately insured and uninsured counterparts. Medicaid insurance status was a significant, independent predictor of death after three years—even after controlling for other clinical factors that could increase someone's risk of poor outcomes.

This should sound familiar to Reason readers: I’ve made similar observations about Medicaid’s health outcomes before, and I think it’s important to look at these studies in the context of Medicaid’s fiscal effects on state and federal government, as well as the recent health care overhaul’s planned expansion of the program: Already the program is wrecking state budgets, but state officials don’t have much flexibility to adjust the program to fit the needs of their states. Instead, thanks to the program’s insistence on matching federal dollars to state spending, they do have an incentive to continually increase the size of the program during times of economic growth. According to a Congressional Budget Office report released this week, block granting the program—as many cash-strapped governors have requested, but as the Obama administration has so far refused to consider—would reduce Medicaid’s cost by an estimated $287 billion over the next decade.

Meanwhile, thanks to the expansion of eligibility called for in the health care law, approximately 16 million additional individuals—and perhaps millions more than that—are expected to end up on Medicaid’s rolls over the next decade. That represents a huge portion of the law’s near-trillion dollar first-decade expense. But if Medicaid’s health benefits are as dubious as these studies seem to suggest, then we’ll be spending a lot of money for a very little in terms of health benefits.

Comparative judgments based on studies like these can be difficult to make with absolute certainty; the research isn't perfect. In particular these studies have a hard time controlling for the negative health effects associated with poverty. And studies do tend to show somewhat greater health effects at the lower end of the income spectrum (which suggests that we should be trimming eligibility rather than expanding it up the income scale). But as Avik Roy has documented in extensive detail, the evidence that, overall, Medicaid coverage correlates with poor health outcomes is fairly strong across numerous studies. How long must we continue to invest hundreds of billions of dollars into a program of such dubious value?