In 2008, Oregon expanded its Medicaid coverage using a lottery. Two years later, the state had a treasure trove of data – blood-pressure, cholesterol, and glycated hemoglobin levels; depression screens; medication inventories; and self-reported information about health status, diagnoses, care utilization, and out-of-pocket costs for care.
The study found that, for those newly covered by Medicaid, the coverage alone did not have a significant on hypertension, cholesterol, or medications used to treat those conditions. There was a significant increase in the diagnosis of diabetes and in the use of mediations for diabetes, but no difference in an important measure of diabetes. These findings led anti-Medicaid expansion pundits to say that the data proved that Medicaid coverage – and perhaps any insurance coverage – doesn’t improve health care.
But in fact there were three significant effects of Medicaid coverage: a significant decrease in depression, a significant increase in the use of preventive services, and a virtual elimination of catastrophic costs of medical care. There were other, less quantifiable effects as well, including the fact that more people took charge of their own medical conditions.
The lottery itself involved 90,000 people that were on a waiting list for Medicaid. From these, some 30,000 were drawn. Of the 12,000 people in the study, which is published in the May 2, 2013 issue of the New England Journal of Medicine, half received Medicaid coverage through the lottery while the other half were among those not selected.
Medicaid coverage resulted in a 30 percent decrease in depression. A positive result on screening for depression was defined as a score of 10 or more on the PHQ-8 (which ranges from 0 to 24, with higher scores indicating more symptoms of depression). There was no increase in the use of medication for depression.
Depression was the most prevalent of the four conditions looked at, and the authors said it was difficult to detect changes in health because of the small numbers of patients with hypertension, high cholesterol, and diabetes.
The study also found a significant decrease in financial stress. The authors noted that health insurance “is a financial product that is aimed at providing financial security” by protecting people from the costs of being injured or sick – and by insuring that providers are paid.
“The results confirm that Medicaid coverage increased overall health care utilization, improved self-reported health, and reduced financial strain,” the authors conclude.
We asked lead author Katherine Baicker, Ph.D., of the Harvard School of Public Health, about the connection between Medicaid coverage and reduced depression. “We saw an increase in diagnosis of depression after the lottery – meaning that more cases were discovered or diagnosed by health care providers in the lottery winners than in the control group,” she told Behavioral Healthcare.
An increase in diagnosed depression, but a decrease in depression? Here’s how that works: “The increase in post-lottery diagnosis means that more subjects in the treatment group than the control group said that a health care professional had given them a diagnosis of depression,” said Baicker. “Our clinical assessment of depression was based on administering a common series of questions (PHQ-8) used to assess depression.”
Fewer people with Medicaid scored in the depression range of the PHQ-8 than people in the control group, she said. This is consistent with a similar share of the Medicaid and control groups suffering from depression before the lottery, she said. After the lottery, a greater share of the Medicaid group then went to a provider who diagnosed their condition. By the time of the assessment, there was greater improvement in the mental health of the Medicaid group than the control group, seen in “lower rates of observed depression,” Baicker explained.
Asked whether the decrease in financial strain resulting from elimination of medical costs was linked to the decrease in depression, Baicker demurred. “We cannot speak directly to that pathway,” she said, but added, “Participants certainly described significant psychic costs associated with being uninsured.”
The role of medications and therapy in reducing depression was not looked at for this study. “Future work will explore the details of treatment, medication, etc. for depression and other chronic conditions,” said Baicker.
A patient’s view
The Washington Post interviewed Mary Carson, a 55-year-old woman who was accepted into Oregon’s Medicaid program in 2011. Here is Carson’s response to the interviewer’s comment that some people have interpreted the study to show that being covered by health insurance doesn’t really improve health.
“Some people have completely lost track of what health insurance is supposed to be,” responded Carson. “We’re talking about somebody being able to get their broken arm fixed if they fall out of a tree. My blood pressure is still not perfect, but over the last two years I have stopped taking two different blood pressure medicines and am only taking half of a third. That is a health improvement but it doesn’t necessary show up in the study. My blood sugar is not perfect, but it’s more consistently in the right zone. But according to the study, I haven’t improved.