In recent years, the narrative surrounding all manner of policy issues in American political debate has often boiled down to how much money the government is (or should) be spending on social or other programs. Nowhere has this been truer than when it comes to health care. A recent study published in Health Affairs, one of the top health policy journals in the country, took a look at Medicaid costs in the 10 highest- and lowest-spending states in the U.S. to try to get at the differences in care as we move into a period of Medicaid expansion mandated by the Affordable Care Act. Authors Todd P. Gilmer and Richard G. Kronick examined not just how much these states were paying out per capita, but on which services that money was spent. Their results tell an interesting story.
It’s clear to someone with even the most basic understanding of the Medicaid system that given the wide discretion states are permitted, different states implement the program differently, spending wildly different amounts of money in the process. But Gilmer and Kronick pose the following question:
Do states that spend significantly more or less than the national average per beneficiary do so because they spend more or less than the national average for a unit of service, or because they provide more or fewer services per person than average?
In other words, are high-expenditure Medicaid states like Arkansas shelling out more money because they’re paying more for the same services as low-expenditure states like Oklahoma? Or is it because they’re providing more or different types of care? This is a fascinating question that has the potential to answer a lot about the “how much” debate that rages on all over the country.
Between 2001 and 2005, most states spent between $7,547 and $10,066 per capita on the type of Medicaid patients examined in the study.* Gilmer and Kronick found that, for the most part, increased spending was due to increased volume of services provided. Approximately 72% of the increase in the 10 highest-spending states was due to providing additional services, with the remaining amount due to price and other effects. But there were huge differences among the highest and lowest 10 in the specific types of services that Medicaid covered.
The most serious difference between the two groups of states is the amount expended on inpatient care. The average amount expended per beneficiary was much higher for those admitted to hospitals than for those who utilized outpatient services. Regionally, New England in particular allocated a much higher percentage of its Medicaid expenditures on outpatient care than inpatient care. And an increased supply of primary care physicians in Washington State correlated positively with reduced hospital admissions rates for chronic illnesses like diabetes, asthma, and heart disease.
So what does this all mean? When states talk about cutting their Medicaid spending, often that reduced funding will, in practice, necessitate reduced numbers of services provided. You can only cut costs for services so much. But these results also clearly point to preventive options being one successful way to reduce costs. By focusing on ongoing maintenance care for Medicaid recipients with chronic illnesses instead of inpatient responses, high-spending states can reduce their costs significantly.
Are these results shocking? Absolutely not. Proponents of health care reform have been advocating this point for a long time. But it’s yet another study with hard facts demonstrating that the primary care model saves the Medicaid system – and in the end, the taxpayers – money in the long run.
*For those interested in the research design, Gilmer and Kronick used data from the Medicaid Analytic eXtract system from 2001-2005, the most complete information available at the time. They looked at Medicaid-only beneficiaries to control for missing data in Medicare expenditures for those who dual-qualify. They also limited the participants to beneficiaries who are disabled, as they are the only group with national Medicaid eligibility. (Eligibility based on other factors varies from state to state.) They also limited their scope to fee-for-service, again because data were not available for managed care payments, and to acute care services as a unit of measurement.