The New York Times fails to explain: Out in Oregon, low-income people who lacked Medicaid coverage got thrown into a lottery.
The state had enough extra money to extend coverage to some, not to all. On this basis, some people from the lottery pool received Medicaid coverage. Other people just like them did not.
This created the opportunity for the Oregon Medicaid study, whose results are now being used to convince the world that Medicaid coverage is worthless, just no damn good. In her news report in the New York Times, this is the way Annie Lowrey described the effects of receiving coverage:
LOWREY (5/2/13): [The Oregon study] found that those who gained Medicaid coverage spent more on health care, making more visits to doctors and trips to the hospital. But the study suggests that Medicaid coverage did not make those adults much healthier, at least within the two-year time frame of the research, judging by their blood pressure, blood sugar and other measures. It did, however, substantially reduce the incidence of depression, and it made them vastly more financially secure.According to Lowrey, Medicaid coverage didn’t make enrollees much healthier, although it lowered their rates of depression and made them more secure financially.
Depression is a major health problem; in this study, Medicaid coverage lowered its incidence by 30 percent. That is a major gain in a major health outcome. For that reason, the taxonomy Lowrey displays in this paragraph makes little sense from the start.
Having said that, let’s consider the matter of financial security. This study does not compare people who got health care through Medicaid to people who got no health care at all. Presumably, many people who lost this lottery continued to pay for their own health care. They may have received just as much health care as the lottery winners, except they had to pay the full fare by themselves.
This may explain their lower rates of financial security and their higher rates of depression. In such matched cases, Medicaid is very much working, even if other health outcomes are exactly the same. (The lottery winners don’t have to go broke, and become depressed, to get their blood pressure medication.)
Still and all, Lowrey said, right out of the gate, that Medicaid coverage didn’t make “those adults much healthier...judging by their blood pressure, blood sugar and other measures.” Later, she tried to go into more detail about this apparent failure.
When she did, she encountered a very important term. That key word was “significantly,” which later seemed to morph into “noticeably:”
LOWREY: The researchers found that Medicaid coverage did not significantly affect the prevalence or diagnosis of hypertension or high cholesterol, or the use of drugs used to treat those conditions. It significantly increased the probability that a person would receive a diagnosis of diabetes and be treated, though it did not reduce blood sugar levels noticeably.Lowrey will never be depressed because she lacks medical coverage or because she had to go broke paying for her health care. Having said that, let’s note the way she tried to report the changes in the prevalence of conditions like hypertension and high cholesterol.
Where Medicaid seemed to have the strongest measured impact was on depression. Getting Medicaid coverage reduced the probability of a positive screening by more than 30 percent.
“Medicaid coverage did not significantly affect the prevalence or diagnosis of hypertension or high cholesterol,” Lowrey writes. This is odd, because as we can see from this Kevin Drum post, the prevalence of high cholesterol was actually 17 percent lower in the group which had Medicaid coverage.
Also from Drum, “the incidence of high glycated hemoglobin levels (a marker of diabetes)” was 18 percent lower among those with Medicaid coverage.
Those results may not seem “significant” to Lowrey, but they may seem significant to the people who won the Medicaid coverage, and perhaps to the ratty people who belong to their lower-class families. Fine ladies may not spend much time thinking about such lesser beings. Instead, they toss off claims about “significance,” perhaps without knowing that they are interacting with the artefact known as “statistical significance.”
Significance and statistical significance simply aren't the same thing. It only gets worse when Lowrey, or perhaps her even less capable editor, uses “noticeably” in place of “significantly,” presumably to vary the patter, this improving the Times reader’s aesthetic pleasure.
Here at THE HOWLER, we don’t really understand how the concept of statistical significance works in a study like this one. That said, we’d damn sure find out before we wrote a news report for the New York Times saying what Lowrey’s piece did.
Four days after Lowrey’s failure, New York Times columnist Ross Douthat followed suit, saying this about the Oregon study: “With a few modest exceptions, the level of insurance had no significant effect on the participants’ actual wellness.” In fact, the Medicaid recipients had an 18 percent lower rate of high glycated hemoglobin levels and a 17 percent lower rate of high cholesterol. Such effects may seem very significant to the people involved.
Lowrey and Douthat are both Harvard grads. The school is turning out some awful young people, and the New York Times can’t wait to employ them, just as it loves to print the bungled op-ed columns of high-ranking Harvard professors. Increasingly, this is the shape of the guild which poses as our upper-end press corps.
Final suggestion: Click here to watch Harvard students responding to Professor Sandel in the first lecture of his life-changing “Justice” course. In particular, watch the young people tell the world if cannibalism was permissible among a small set of British sailors who were shipwrecked many long years ago.
It never occurs to these young people that they have never faced any experience which dimly resembles the experience under review. If you watch, you will see some very young, inexperienced people who seem to have a very high sense of self regard. You will also see a famous professor pandering to their presumed greatness and thus to their youthful delusions.
Ten years later, people like these will be too lazy to learn their craft when they're employed at the Times. Instead, they will tell Times readers that it isn’t significant or noticeable when low-income people seem to get better health results.
There’s more to be said about “statistical significance” and the role the term plays in this latest gong-show, in which people are being sold the latest conservative line. That said, we see no sign that Lowrey understood what she was saying in that news report.
Why in the world is she in that job? And who in the world was her editor? Final question:
Will Rachel Maddow lower the boom on this gong-show tonight?