When Rudy Giuliani was running for president he cited data comparing healthcare in England and the United States. This is the essence of what he reported:
In Britain about 44% of men diagnosed with prostate cancer were alive 5 years later. On the other hand, in the US about 81% men diagnosed with prostate cancer were alive 5 years later. So Giuliani concluded that the 5-year survival rate for prostate cancer in the US is nearly double what it is in England (44% for England vs. 82% for the US). He was wrong. The risk of death from prostate cancer is virtually the same in both countries.
In the US most prostate cancer screening involves looking for prostate-specific antigens (PSA). British doctors do not rely on PSA testing near as much as their American counterparts. In England most prostate cancer is identified through symptoms. These different approaches to diagnosing prostate cancer are key.
Here is how different screening practices lead to different survival rates.
The US has a higher proportion of prostate cancer diagnoses because US physicians rely on PSA screening. PSA testing often identifies small tumors that grow so slowly they do not harm a man before he dies from natural causes due to old age. These are sometimes called nonprogressive cancers. Nonprogressive cancers meet the pathological definition of cancer but never cause symptoms during a patient’s lifetime. On the other hand, because British physicians rely more on symptoms to diagnose prostate cancer, they usually do not diagnose men with nonprogressive cancers.
So if PSA testing in the US identifies 3000 men with prostate cancer, 2000 of those may have nonprogressive cancer while 1000 may have progressive cancer. The 2000 nonprogressive cancer patients survive along with 440 of the 1000 progressive cancer patients. Thus the survival rate is 2440/3000 = 81%
If England relies on symptoms to identify prostate cancer, then it will miss the 2000 with nonprogressive cancer and just diagnose the 1000 with progressive cancer. Among the 1000 progressive cancer patients, 440 survive, just like in America. However, England’s survival rate is 440/1000 = 44%.
The upshot is that over diagnosing prostate cancers with PSA testing in the US is inflating survival rates. In reality, the risk of dying from prostate cancer is virtually the same in both countries. A 2001 report showed that there were 26 prostate cancer deaths per 100,000 men in the US and 27 cancer deaths per 100,000 men in England (Shibata & Whittemore).
The other dark secret of PSA testing is that it may be leading to unnecessary treatment. If PSA identifies a tumor then a man may choose surgery and/or radiation therapy. These treatments may be unnecessary if the tumor is nonprogressive, yet the man could be left with impotence and incontinence as a consequence of being treated.
If PSA testing identifies a cancer tumor, I think the next important question should be: "Is it progressive?" The answer(s) to this question and a thorough discussion between patient and doctor should inform one's decision to withhold or undergo treatment. (Source: Gigerenzer et al., 2008. Helping Doctors and Patients Make Sense of Health Statistics)
In Britain about 44% of men diagnosed with prostate cancer were alive 5 years later. On the other hand, in the US about 81% men diagnosed with prostate cancer were alive 5 years later. So Giuliani concluded that the 5-year survival rate for prostate cancer in the US is nearly double what it is in England (44% for England vs. 82% for the US). He was wrong. The risk of death from prostate cancer is virtually the same in both countries.
In the US most prostate cancer screening involves looking for prostate-specific antigens (PSA). British doctors do not rely on PSA testing near as much as their American counterparts. In England most prostate cancer is identified through symptoms. These different approaches to diagnosing prostate cancer are key.
Here is how different screening practices lead to different survival rates.
The US has a higher proportion of prostate cancer diagnoses because US physicians rely on PSA screening. PSA testing often identifies small tumors that grow so slowly they do not harm a man before he dies from natural causes due to old age. These are sometimes called nonprogressive cancers. Nonprogressive cancers meet the pathological definition of cancer but never cause symptoms during a patient’s lifetime. On the other hand, because British physicians rely more on symptoms to diagnose prostate cancer, they usually do not diagnose men with nonprogressive cancers.
So if PSA testing in the US identifies 3000 men with prostate cancer, 2000 of those may have nonprogressive cancer while 1000 may have progressive cancer. The 2000 nonprogressive cancer patients survive along with 440 of the 1000 progressive cancer patients. Thus the survival rate is 2440/3000 = 81%
If England relies on symptoms to identify prostate cancer, then it will miss the 2000 with nonprogressive cancer and just diagnose the 1000 with progressive cancer. Among the 1000 progressive cancer patients, 440 survive, just like in America. However, England’s survival rate is 440/1000 = 44%.
The upshot is that over diagnosing prostate cancers with PSA testing in the US is inflating survival rates. In reality, the risk of dying from prostate cancer is virtually the same in both countries. A 2001 report showed that there were 26 prostate cancer deaths per 100,000 men in the US and 27 cancer deaths per 100,000 men in England (Shibata & Whittemore).
The other dark secret of PSA testing is that it may be leading to unnecessary treatment. If PSA identifies a tumor then a man may choose surgery and/or radiation therapy. These treatments may be unnecessary if the tumor is nonprogressive, yet the man could be left with impotence and incontinence as a consequence of being treated.
If PSA testing identifies a cancer tumor, I think the next important question should be: "Is it progressive?" The answer(s) to this question and a thorough discussion between patient and doctor should inform one's decision to withhold or undergo treatment. (Source: Gigerenzer et al., 2008. Helping Doctors and Patients Make Sense of Health Statistics)
Continue reading at the original source →