Last week, the often controversial and always conservative William Saletan published a piece about the link between race and cancer outcomes. In what's sure to give black folks flashbacks of the Tuskegee Experiment and "The Bell Curve," Saletan concludes that genetic differences between races largely explain a study that found a significantly lower survival rate among African Americans with certain cancers. It's a tenuous argument, not to mention a loose interpretation of the report.

The study, published by the National Cancer Institute, reveals that African Americans are 21 to 61 percent more likely to die from ovarian, breast, and prostate cancers than whites, but that factors such as income, education, and access to medical care aren't to blame. That pretty much rules out everything except genetic makeup-the likely culprit, according to Saletan's interpretation of the data:

The problem with this reasoning is that any genetic scientist with a basic understanding of the human genome will tell you that, when it comes to race, there is no cracking the DNA code. There's no gene that equals black, or white, or Asian. Furthermore, this focus on race ignores other weighty contributors to poor cancer outcomes, such as family history or the role obesity plays in increasing our overall risk. At its most disturbing, categorizing humans based on skin color reeks of the notion of genetic inferiority.

And as if on cue, Saletan's theory takes a turn down this road when he attempts to bolster his position by referencing a story he wrote about heredity, racial gaps, and IQ tests, which claims that white American kids have bigger brains that black American kids. He goes on to argue that people are afraid to acknowledge genetic differences along racial lines for fear "we'll lay a scientific foundation for segregation."

Uh, yeah. It's been done before. The idea of racial hierarchy was the justification for enslaving millions of Africans and the mass slaughtering of Jews.

And herein lies the problem: Based on our collective history of dealing with race, it's not hard to understand why presenting a case for genetic coding and ethnicity incites fear. Once we start using DNA as a way to determine a propensity for disease, we leave ourselves open to a wide interpretation of how that information could be used. And the results could have some serious medical consequences.

Would insurance companies, for example, increase their rates for black Americans on the basis that we're predisposed to pre-existing conditions? As a 33-year-old African American ovarian cancer survivor, that kind of medical determination is already something I worry about.

It's a shame really, because what gets lost in Saletan's analysis is that there is real value in studying the health trends of certain populations. If nothing else, the NCI report demonstrates that a one-size-fits-all approach to treatment doesn't work, which is critical to identifying more-effective therapies for African Americans and other ethnic groups. But thinking of race as "a scientific category" only drudges up bad memories and offers nothing new to help narrow the survival gap.

--CHANA GARCIA