Wednesday, October 12, 2016

Journalistic trafficking in cliches and stereotypes

An interesting, though challenged article, Your Surgeon Is Probably a Republican, Your Psychiatrist Probably a Democrat by Margot Sanger-Katz.
We know that Americans are increasingly sorting themselves by political affiliation into friendships, even into neighborhoods. Something similar seems to be happening with doctors and their various specialties.

New data show that, in certain medical fields, large majorities of physicians tend to share the political leanings of their colleagues, and a study suggests ideology could affect some treatment recommendations. In surgery, anesthesiology and urology, for example, around two-thirds of doctors who have registered a political affiliation are Republicans. In infectious disease medicine, psychiatry and pediatrics, more than two-thirds are Democrats.

The conclusions are drawn from data compiled by researchers at Yale. They joined two large public data sets, one listing every doctor in the United States and another containing the party registration of every voter in 29 states.
As usual, the two most important questions, "What are the effect sizes?" and "Why is this important?" are not directly answered. Journalists generally function, intentionally or unconsciously, as an advocacy group for one particular world view. That world view might be harshly described as Reform Marxist/SJW/Far Left. Or it might, more generously, be described, and perhaps more accurately, as the world view of insulated, well intentioned, upper middle class, high income, highly educated in the humanities (generally suffering some degree of innumeracy), urban, white-collar professionals. Not bad, but generally lacking perspective and context with a proclivity to virtue signaling and overconfidence to the point of unsupportable certainty.

Americans tend to self-identify as 40% conservative, 40% moderate and 20% liberal (party affiliation does not necessarily match political orientation).

The first question - "What are the effect sizes?" Hard to tell from the reporting. They filter based on Republican and Democrat. How large is the group in the middle? If only 10% are registered Republican/Democrat and everyone else is Moderate, then there would appear to not be much of interest in this study because it would be reporting on the fringes.

The reporter leaves it to the reader to do the heavy lifting. A footer indicates that:
But over all, the researchers were able to collect complete data for more than 55,000 physicians living in the 29 states where voter files include party registration.

There is a graphic with an important rider:

Click to enlarge.

The rider is: "Data from a sample of 34,532 physicians in 29 states." In other words, 63% of physicians were registered to vote by party affiliation and 37% either were not registered to vote or did not have a party affiliation.

The fact that we are talking about 63% of physicians and not all physicians is not a disqualifying issue, but it is an important caveat to know in order to understand and interpret the data correctly.

What about the second question, "Why is this important?" It is not clear that party affiliation is important at all. There seems to be no material impact on outcomes.
“These findings suggest you are going to get different care,” Professor Hersh said, adding that the differences might not matter much for the average patient. But they might for patients whose needs were closely related to politically divisive subjects, like reproductive health, with issues like contraception, abortion and prenatal screening; or H.I.V. prevention, with risk factors that include sex and intravenous drug use.

Primary care doctors and obstetrician-gynecologists, the doctors most likely to consider such issues, were among the most evenly split in the study sample. That means that patients probably can’t guess the political leanings of their doctor without asking (or checking the voter file data). The current study is only a survey, but Professor Hersh said he hopes the research spurs more examinations of how ideology shapes medical practice.
I am glad that Margot Sanger-Katz wrote this article as it provides some marginally interesting information but it certainly falls far short of what it could have been.

How can you write about political affiliation in the hospital setting without retelling the marvelous story of Ronald Reagan as he entered surgery after the assassination attempt on him:
3:24 p.m. Reagan was wheeled into the operating room. He had lost about 2,100 cc of blood, but his bleeding had slowed and he had received 4 1/2 replacement units. As he was moved from the stretcher to the operating table, he looked around and said, "Please tell me you're all Republicans." Giordano, a liberal Democrat, said, "We're all Republicans today."
I observe that the commenters, many of whom can't resist the temptation of pop psychologizing based on party affiliation, partisan sniping and committing the fundamental attribution error, and otherwise clogging the system with trivia, also generate some quite worthwhile content and observations.

Clare A observes that the study left out other independent variables which might contribute to party affiliation separate from field:
I wish I could say that this is malarkey but being a med student and the kid of two physicians, I know better than to do so. Though I will say this: geography has an astounding influence on the political positions of those in the medical field. Having worked in both Detroit and Birmingham (an affluent suburb outside Detroit), I can tell you that the surgeons in Detroit, even the ones from the wealthy/conservative deep south, are far more left leaning than the ones in Birmingham. Even my father, who was raised on food stamps and programs supported by the democratic party, has become alarmingly more conservative since working in an affluent area. Then again, my mother who is an ER physician and was raised in the 1% is as left leaning as they come, despite the odds. So who really knows!
Orthodoc suggests that this is much ado about nothing.
Interesting, trend in the comments, following many tired old judgments: surgeons etc want money, want action, and are impatient or biased. I am an orthopedic surgeon. My wife is also in a surgical specialty. I am/was not motivated by "risk/action", or particularly by money. I am a Democrat. I think the graph most telling is the one looking at age. I suspect that older white male doctors are more likely to be republican, mirroring the rest of the country. Surgery has traditionally been more male dominated. While this is changing with more female surgeons, there are still more males than females. This is partly due to the demands and hours required by the profession as well as from the slow change inherent in the medical system of training and practice. So, I suspect the political leanings of doctors is more correlation than causation. I would not read too much into trying to buttonhole an individual person into a group based on simple profession. I have met doctors, lawyers and blue collar workers of all political stripes and with all ranges of personal motivations.
Steve lambasts Sanger-Katz for her lazy stereotyping and statistical incomprehension:
The conclusions proposed by the author are an offensive extrapolation beyond the data presented in this article. While the trends of party affiliation presented in this article may be true, the "potential explanations" presented by the author revolve around age-old, biased stereotypes of physicians: that those who practice internal medicine and related subspecialties are the doctors with heart, while surgeons remain the brutish, egomaniacal, self-serving group they have always been.

In short, the author infers throughout her argument that surgeons simply don't care about patients as much as other doctors.

This is derogatory, untrue, and dangerous. I am a proud Democrat and a proud surgeon. Care, compassion, and belief in the absolute necessity of a strong social and medical safety net are not isolated to particular fields of medicine located outside the OR - nor to a particular political party for that matter. To infer otherwise shows a lack of editorial restraint and undermines the care that hardworking, compassionate surgeons tirelessly provide on a daily basis.
Tony Longo adds some humor but makes an important point.
I can hone this further very easily: my psychiatrist may be a run-of-the-mill Dem, but my actual weekly Therapist is a highly motivated leftist with fire in her gut. My internist is a very beneficent dude, but I suspect my orthopedist of certain semi-conservative leanings….In dentistry, the main man’s an old-fashioned socialist, when he remembers to, but the root canal guy has some weird ideas, man, and Ayn Rand on the coffee table. Dermatology, quite libby; ENT, so authoritarian I’m afraid to bring it up.

As for my chiropractor, if you can interrupt the Barbra Streisand record he’s listening to at the moment I think he’d offer some rather mellow and advanced views on things in general.

Don't even ask about my last nutrionist.

The mainstream media seem bent on fostering partisanship, division, hysteria, and speculation unmoored on known facts.

I wish they would serve us better by going beyond the cliche to deep thought.

No comments:

Post a Comment