Tuesday, February 19, 2008

Prognosis is Poor Without Health Insurance

A study reports on something which is also the common-sense conclusion to come to: If you have no health insurance and little money you are less likely to see a doctor regularly or early when symptoms appear. So you go and seek treatment much later than would be good for you:

A nationwide study has found that the uninsured and those covered by Medicaid are more likely than those with private insurance to receive a diagnosis of cancer in late stages, often diminishing their chances of survival.

The study by researchers with the American Cancer Society also found that blacks had a higher risk of late diagnosis, even after accounting for their disproportionately high rates of being uninsured and underinsured. The study's authors speculated that the disparity might be caused by a lack of health literacy and an inadequate supply of providers in minority communities. The study is to be published online Monday in The Lancet Oncology.

Previous studies have shown a correlation between insurance status and the stage of diagnosis for particular cancers. The new research is the first to examine a dozen major cancer types and to do so nationally with the most current data. It mined the National Cancer Data Base, which began collecting information about insurance in the late 1990s, to analyze 3.7 million patients who received diagnoses from 1998 to 2004.


When comparing blacks to whites, the disparities in late-stage diagnosis were statistically significant for 10 of the 12 cancers. Hispanics also had a higher risk but less so than blacks.

The study's authors concluded that "individuals without private insurance are not receiving optimum care in terms of cancer screening or timely diagnosis and follow-up with health care providers." Advanced-stage diagnosis, they wrote, "leads to increased morbidity, decreased quality of life and survival and, often, increased costs."

The study cites previous research that shows patients receiving a diagnosis of colon cancer in Stage I have a five-year survival rate of 93 percent, compared with 44 percent at Stage III and 8 percent at Stage IV.

"There's evidence that not having insurance increases suffering," said Dr. Otis W. Brawley, the American Cancer Society's chief medical officer.

Note that not having health insurance has an impact on continuity of care, but that being black or Hispanic has an extra impact on top of that. The reasons for the extra impact might be the ones the quote suggests. My guess would be that it has to do with lack of doctors in overwhelmingly low-income areas. Such areas are also often racially or ethnically segregated. In short, a white uninsured person with no money might live closer to sources of regular medical care because of less housing segregation, whether voluntary or not.

Despite the importance of this article, I have trouble with the way screening is sold here. Consider some disease that can be spotted by screening and suppose, just as a thought experiment, that early treatment actually has no benefit at all. Suppose also that the disease takes five years to progress from Stage I to Stage IV and then one year to kill. Given this, people who are diagnosed by screening in Stage I would live an extra six years, people who are diagnosed by advanced symptoms in Stage IV would live one year. But in reality all these people are living exactly the same length of time with the disease; it's just that some of them know they have it earlier.

This is why I'm not too satisfied with the use of the five-year survival rates as a measure of how good the treatment is.

Of course the above is just a thought experiment, and no way am I saying that early detection has no benefits. It obviously does, in every case where early treatment improves the final outcome. But screening in itself is not a treatment.