Patients diagnosed by urologists specializing in radiotherapy, cryotherapy or external beam radiation therapy were more likely to get treated with—you guessed it—radiotherapy, cryotherapy, or external beam radiation therapy, a study shows.
To the hammer, all the world is a nail.
And to the surgeon, all patient care is a scalpel, or so the saying goes.
And so it is for urologists, those doctors who diagnose low-risk prostate cancer in men with a life expectancy of less than 10 years, a patient group that should be managed with observation, not treatment, guidelines recommend.
Men diagnosed by urologists whose claims history indicates a preponderance of prostatectomy procedures were more likely to get a prostatectomy than any other form of care including watchful waiting.
Likewise, patients diagnosed by urologists specializing in radiotherapy or cryotherapy or external beam radiation therapy were more likely to get treated with—you guessed it—radiotherapy, cryotherapy, or external beam radiation therapy.
Yet these were all patients diagnosed with the same extremely low severity of disease.
These were the not-so-surprising but important take-aways this week from a study in JAMA Internal Medicine by Karen Hoffman, MD, a radiation oncologist at MD Anderson Cancer Center in Houston. Her look at a large national sample of men diagnosed with low-risk prostate cancer reveals a crying need for greater transparency in physicians' practice biases.
Low-risk prostate cancer was defined as tumors classified as stage T1 to T2a, a Gleason of 6 or less, and a serum prostate specific antigen level that was less than 10.
Not only do treatment risks such as urinary dysfunction, rectal bleeding, and impotence far outweigh benefits, but evidence now shows that low-risk prostate cancer is unlikely to cause symptoms or progress if left untreated.
"Our study does suggest that physicians who diagnose prostate cancer have some bias in the way they treat patients, a bias toward approaches that they themselves use," Hoffman says.
While her study can't say exactly why these men got the treatments they did, she says "it's possible that financial incentives play a role, because these upfront treatments provide more reimbursement to these physicians."
Hoffman adds that, "it's certainly fair to say that men with less than a 10-year life expectancy and low-risk prostate cancer are not expected to die from their disease. I'd expect the rate of death from prostate cancer for these men would be less than 1%."
Nearly all men in their late 70s and 80s with low-risk disease who undergo prostate cancer treatment are inappropriately treated, she says.
Were these men told they could easily go without treatment? We don't know. Maybe they had fewer than 10 years of expected life, but the doctor couldn't bring himself to bring that into the conversation for fear of appearing to ration care or upset the wife.
Hoffman emphasizes that her findings point out a major quality issue for cancer patients, who have no way of knowing how much their doctors' treatment preferences or financial incentives influence their recommended course of care.
I've known of physicians who, despite evidence that certain laparoscopic procedures were more effective and safer, continued to perform open surgical procedures because they'd never done the required laparoscopic training.
Likewise for some doctors who won't administer clot busting drugs to patients with ischemic stroke. Many don't do it because they're afraid and don't have the skill to know which patients will benefit.
Hoffman confirms it's not just prostate cancer care that's influenced by physician bias.
She pointed me to a study published last month in the Journal of Clinical Oncology showing a worrisome trend in how some physicians diagnose breast cancer with marked regional variation.
Women with suspected breast disease were more likely to be diagnosed with the use of more invasive surgical excisional biopsy than what guidelines now recommend, the needle biopsy. Factors influencing the more aggressive strategy included lack of board certification, training in a U.S. residency program, recency of training, whether the physician was a surgical oncologist, and whether the practice specialized in breast cancer care.
Hoffman says patients and referring physicians have no way to know what their physicians' preferences and proclivities might be, whether they're more or less aggressive, their treatment history, and their outcomes like complication rates and survival.
But they should.
She suggests some public reporting of each physician's cancer management "profile" would inform choices. "We don't have that, "she says, "and while knowing whether a physician is likely to manage a patient's cancer with observation is appropriate, it should be balanced with treatment complication rates, and how doctors follow-up and communicate with their patients."
Hoffman acknowledges that for a few men in this large sample of 12,068 older men, treatment rather than observation was the right option. But probably not all that many, she says. It all depends on how long they actually live, and no one has that kind of crystal ball.
And she emphasizes that these urologists were not violating standard of care during the period of this study, which looked at men diagnosed between 2006 and 2009. That was before the National Comprehensive Cancer Network changed its guidelines for prostate cancer care recommending no treatment for patients in this group.
They just recommended the procedure they usually performed.
Hoffman says that her study's findings "resonate across all fields of medicine, because it's about the impact the individual physician the patient sees has on treatment choice."
When a close friend was recently diagnosed with breast cancer, we talked about how to pick an oncologist, a surgeon. What do you look for to learn the doctor's track record.
There is, we realized, no place to go except what other doctors might say, in vague, possibly irrelevant terms.
She was told that surgeon so and so is "nice," and that oncologist so and so is "good."
We want to know, of course, exactly what does that mean?