I have a slight disagreement with Avik Roy on his post from last week on block-granting Medicaid. If you’ve read this blog for any decent length of time you know I’m no fan of this idea, or rather the specific block-granting proposals put forth by the top-tier GOP candidates. First let me say that above all else I appreciate that Roy takes this stuff seriously, even if I disagree with him. So I’ll try to focus on just a few of my disagreements with some of the reasoning behind his stance (all emphasis in bold are mine):
I’ve written often of the very serious problems with the Medicaid program. People on Medicaid have far worse health outcomes than those with private insurance, and in many cases those with no insurance at all. The main cause of Medicaid’s poor outcomes is that it’s very difficult for Medicaid patients to get doctors’ appointments, even for urgent medical problems. This poor access to care, in turn, is caused by the fact that Medicaid severely underpays doctors to care for Medicaid patients. In New York, for example, Medicaid pays 29 percent of what private insurers pay. There are, in essence, two approaches to making Medicaid better. One would be to leave the program as is, unreformed, but throw more money at it, in order to pay doctors more. The other would be to reform the program, so as to make sure Medicaid dollars are directed toward actual health care, instead of waste, fraud, and abuse.
1. People on Medicaid have poorer health outcomes than those A) private insurance or B) no insurance at all. Well, duh to the first:
Which is to say, those who are enrolled in Medicaid are more likely to report fair/poor health status than those with private insurance, are more likely to have chronic conditions, are more likely to be unemployed due to poor health, and thus most of Medicaid’s spending occurs within the disabled and dual-eligible (elderly) categories:
It doesn’t take much to infer that a program that serves people in poor health will have poorer health outcomes than a program that serves people with a better health status. Again, I would imagine that this is primarily a reflection of who the program serves, not of the program itself. All things being equal the poor, elderly, and disabled are all categories that one might suspect as having more health issues than the middle/high income, young and working aged, non-disabled population. Ergo the more health problems a segment of the population has, the greater amount of money it costs to treat and respond to those problems.
Some see this as an issue that block-granting can solve, primarily through the power of Health Savings Accounts. I’ve already touched on the idea that HSA’s are a tool for the sickest among us, and since the largest expenditures in Medicaid occur with the elderly and disabled the rational behind such consumer-directed health spending reforms seems even more misguided.
As to the second part, that people are better off uninsured than on Medicaid, it just isn’t true. Despite being based on erroneous assumption Roy has nevertheless gone so far as to describe Medicaid as a “humanitarian disaster” in the past. The problem is that he’s basing that opinion on studies that, while having shown correlations between poor health and Medicaid enrollment (I.e., what I discussed above), don’t actually prove causality. Furthermore, there is nothing inherent in block-granting or HSA’s that would change the demographic or physical issues that are characteristic of current Medicaid enrollees. This isn’t to say that Medicaid couldn’t be better, or that Medicaid doesn’t have issues, but it seems disingenuous on so many levels to imagine that it’s worse than nothing at all.
2. Roy claims that the main cause for poor health outcomes in Medicaid is due to limited access to doctors, and that this is a result of low physician reimbursement rates. I’ve actually written about this before and remain skeptical. Roy’s evidence for causality in this post are links to two previously written pieces about physicians not accepting new Medicaid patients or appointments. The problem I have with using physician acceptance rates, again, is that it doesn’t necessarily tell us that Medicaid causes poor health outcomes. It also strikes me as a glorified, supply-side, inference that doesn’t tell us the whole story on access. When people (i.e., the demand side) are asked about their access to care issues, those enrolled in Medicaid are leagues ahead of the uninsured and not that far behind the privately insured:
If the end-result of some physicians not accepting new Medicaid patients is supposed to create large gaps in access, it doesn’t show up for those who report “no usual source of care.”
Now it is true that Medicaid pays doctors less than private insurers (chart via another Roy post):
It’s also true enough, looking at state-level data, that higher reimbursement rates tend to lead to higher acceptance rates:
3. Which actually leads to the third and final point. Roy presents an odd, and decidedly false, choice. He writes that we can either leave the program as it is and simply throw money at it, or we can “reform the program, so as to make sure Medicaid dollars are directed toward actual health care, instead of waste, fraud, and abuse.” When I first read this I was confused – I thought the greatest shortcoming of the program centered on reimbursement rates. Here it seems to be the perennially vacuous ‘waste etc.” If the main problem with Medicaid is low physician reimbursement rates, then I would say let’s raise those rates. If we can make the greatest impact in the program, right now, by raising the amount that we pay doctors to see low-income Americans then let’s do it. Indeed that is what the Affordable Care Act does, albeit temporarily, for 2013 and 2014. So what ‘waste’ is Roy referring to, and how does it affect his identification of low physician rates as the primary problem affecting Medicaid?
Fortunately he obliged my inquisitiveness on Twitter, responding that without eliminating waste there could be no extra money to pay doctors. My skepticism aside, he pointed me to an article he wrote in 2010 on how to “fix Medicaid.” Based off of a report on New York, Roy identifies the two biggest problems in that state’s approach to Medicaid (by which I assume he means “rampant waste,” since those are his words and this is the article he points me towards):
In his 17-page report, Ravitch traces the recent history of how New York’s Medicaid program came to this pass. Most important, New York has been one of the most aggressive states in taking advantage of federal matching funds in order to expand its Medicaid program, making it one of the most lavish in the country. Unfortunately, it has been politically easy to expand Medicaid during good times, but impossible to rein it in during bad times; indeed, during the financial crisis of 2008–09, the state actually expanded its Medicaid coverage, and Medicaid enrollment increased by 600,000.
Another problem, the transformation of Medicaid from a welfare program to an entitlement program, was a result of the passage of federal welfare reform in 1996. When Medicaid was first instituted in 1965, nearly everyone eligible for Medicaid was already on welfare: that is, they were receiving direct cash assistance. After 1996, New York’s welfare rolls shrank dramatically, while its Medicaid rolls continued to expand. “Today,” writes Ravitch, “only one out of six New York children and adults receiving Medicaid services also receives cash assistance,” because most people on Medicaid are employed, albeit with below-average incomes.
Hence my continuing, and I believe entirely reasonable, doubt on the motives of block-grant advocates. One cannot help but notice that the two main examples of “waste,” according to Roy, involve more people enrolled in Medicaid and/or enrollment being allowed over 1996 levels. This I believe remains the crux of the argument on block-granting Medicaid – it is not reform aimed at making it a better program that serves those who need it, but instead is reform aimed at offering fewer services, hitting an arbitrary federal budget target, and covering fewer Americans. In effect the “truth is in the pudding,” sort to speak, on the objective of Roy’s preferred reform:
I suppose we’ll just have to disagree on characterizing this aspect of Medicaid that tries to serve those who need it, rather than just those who are lucky enough to get it, as “wasteful.”
Really, there is no reason to construct such a false choice on Medicaid’s future. I believe we can, and should, reform the program without block-granting it into a much more finite resource to the poor. Perhaps we could sustain those reimbursement rates raised in the ACA, which technically matches Medicare reimbursements, work with the states to implement better fraud detection, and finally get serious on a solution to the long term care dilemma. Some of have even suggested federalizing the program altogether. The point is that these are all avenues to reform without kicking millions of people off the Medicaid rolls.