Overdiagnosed is a book written by Dr. H. Gilbert Welch with Dr. Lisa Schwartz and Dr. Steven Woloshin. All three authors are physicians and professors at Dartmouth Medical School. This post has some notes on studies mentioned in the book.
• The number needed to treat – drawn from the results of multiple randomized trials studying hypertension:
Severe (Diastolic BP 115-129): 1.4
Moderate (Diastolic BP 105-114): 4
Mild (Diastolic BP 90-104): 11
Very Mild (Diastolic BP 90-100): 18
• A study of intensive drug therapy to lower blood sugar in type 2 diabetes found that after three years, 5 percent of patients receiving intensive therapy had died, compared with 4 percent of those receiving standard therapy.
• The percentage increase in disease prevalence based on lower diagnostic thresholds:
Diabetes: 14%
Hypertension: 35%
Hyperlipidemia: 86%
Osteoporosis in women: 85%
• The head of the diabetes cutoff panel was paid consultant to multiple pharmaceutical companies which make diabetes drugs.
• Nine of the eleven authors of recent high blood pressure had some kind of financial ties – as paid consultants, paid speakers, or grant recipients – to drug companies that made high blood pressure drugs.
• Eight of the nine experts who lowered the cholesterol cutoff were paid consultants to drug companies making cholesterol drugs.
• The first cutoff for osteoporosis was established by a World Health Organization panel in partnership with the International Osteoporosis Foundation – an organization whose corporate advisory board consisted of thirty-one drug and medical equipment companies.
• Bisphosphonates may actually make bones more brittle by changing the bone architecture. They can also disturb calcium metabolism, lead to ulcers in the esophagus, and, very rarely, cause bone to die.
• In people without any symptoms of gallbladder disease, about 10 percent have gallstones when scanned by ultrasound.
• In people without knee pain or a history of knee injury, about 40 percent have meniscal damage in their knees when scanned by MRI.
• In people without any back pain, over 50 percent have bulging lumbar discs when scanned by MRI.
• If damaged cartilage is the cause of your pain, knee arthroscopic surgery could help, but if arthritis is the cause, that surgery clearly does not help and in fact can only do harm.
• Brain MRIs of over two thousand people – none of whom had had a clinical diagnosis of stroke – found that over 10 percent of these healthy participants had had strokes.
• In a recent study of over a thousand people who elected to undergo total-body CT screening – people with no symptoms – 86 percent had at least one abnormality detected. Because there were so many abnormalities found during the course of the study, more than three thousand abnormalities, the researchers calculated that the average individual had 2.8 abnormalities!
• A study of 201 men between ages sixty and seventy-five with hypertension and/or heart disease detected five aneurysms during physical examinations, while eighteen were detected by ultrasound. Of the thirteen aneurysms detected only by ultrasound, one was large, four were midsized, and eight were small.
• About two hundred thousand Americans are diagnosed with abdominal aortic aneurysms each year, and almost all of the aneurysms – roughly 90 percent – are below the size for which surgery is recommended.
• Studies of trauma patients show that very few develop large clots. But if the trauma patients are examined using duplex ultrasound scans, more than half of them are found to have clots, albeit small ones.
• With a ventilation-perfusion scanning, more than half of patients with blood clots in their legs are found to have small clots in their lungs.
• Spiral CT scans find a third more clots in the lungs of patients with leg blood clots than ventilation-perfusion scans do.
• Most men who die from prostate cancer are elderly; the median age of death is eighty.
• A study that examined prostates from men who had died in accidents who weren’t known to be sick or to have cancer found this prevalence by age group:
ages 20-29: nearly 10%
ages 30-39: 30%
ages 40-49: 40%
ages 50-59: 45%
ages 60-69: nearly 70%
ages 70-79: more than 80%
• If over half of older men have prostate cancer but only 3 percent will ever die of it, the potential for overdiagnosis is enormous.
• A study measured PSA in about ten thousand healthy older men with no evidence of prostate cancer and biopsied all of them regardless of their PSA. Prostate cancer could be found at every PSA level.
• Scientists have begun to uncover biologic mechanisms that halt the progression of cancer. Some cancers outgrow their blood supply and are starved; others are recognized by the host’s immune system and successfully contained; and some are not that aggressive in the first place.
• Roughly 50 percent of men who undergo radical prostatectomy experience sexual dysfunction, a third have problems urinating, and one to two out of a thousand die in the hospital following surgery. About 15 percent of men treated with radiation develop problems with defecation, usually pain or urgency.
• In 2009, the results of two randomized trials that involved over a quarter of a million men were published. The European study concluded that PSA screening reduced prostate cancer mortality by 20 percent, while the study from the USA found that screening increased cancer mortality by 13 percent.
• A study in Finland examined the thyroid glands in 101 consecutive autopsies of older patients who had died in the hospital. Over a third of the autopsied patients had thyroid cancer, and this was a low estimate since many of the cancers were smaller than two millimeters, the width at which they too slices.
• In the SEER study, most of the new thyroid cancer diagnoses are small thyroid cancers, and all of them are papillary thyroid cancer, the least aggressive type.
• Three randomized trials completed in the 1990s showed that screening chest X-rays did not lead to a reduction in lung cancer deaths. In fact, in two of the studies, screening appeared to cause more deaths. There was more lung cancer surgery in the screened groups, and lung cancer surgery itself can cause death.
• Data from 2001 on spiral CT screening on more than five thousand volunteers showed that nonsmokers have about the same risk of lung cancer as smokers. The death from lung cancer (per 1,000 over 5 years) is actually 3.35 for smokers and 0.2 for never-smokers – a ratio of 17.
• Australian researchers estimated that a typical fifteen-year-old girl undergoing regular Pap screening has a greater than 75 percent chance of needing a colposcopy sometime in her lifetime. That’s a lot of overdiagnosis of precancer for a cancer for which the lifetime risk of death is 0.2 percent (two per thousand).
• A large Canadian randomized trial found that mammography and clinical exams did not reduce deaths from breast cancer among women ages forty to forty-nine.
• None of nine randomized trials on mammography showed that mammography led to a reduction in death for younger women.
• According to the SEER data, about 90 percent of women with metastatic breast cancer ultimately die from breast cancer.
• Mammograms miss about one-quarter of cancers that are destined to appear during the following year.
• Long-term follow-up of over 215,000 women who had had normal mammograms in New Mexico showed that their risk of developing cancer over the next seven years was almost exactly that of similar-age women in the general population.
• Based on seven autopsy studies involving more than a thousand women, somewhere between 2 and 40 percent of women who were not known to have breast cancer – and who died for some reason other than breast cancer – nonetheless has pathologic evidence of breast cancer upon examination.
• Some of the cancers found by mammography are not destined to progress to become clinically evident. A substantial portion even appear to regress.
• Small lung nodules are detected on chest CT scans in roughly 15 percent of nonsmokers and in up to 50 percent of smokers. The vast majority of these nodules will never become cancer.
• Chance that an incidentaloma represents a lethal cancer for a typical fifty-year-old:
Lung (smokers): 3.6%
Lung (never-smokers): 0.7%
Kidney: 0.2%
Liver: 0.5%
Thyroid (by ultrasound): <0.01%
• A study of fifty-three kidney tumors found that 14 percent got smaller and 40 percent grew so slowly that it would take more than six years for them to double in size.
• The CAST randomized trial of treating heart rhythm disturbances found that the death rate in treated patients was two and a half times higher than in those given a placebo.
• A review of studies involving more than 37,000 women found that fetal monitoring doesn’t reduce the number of newborns who develop cerebral palsy, have to go to the intensive care unit, or die.
• For every hundred fetuses found to have abnormalities, three at most will actually have trisomy.
• A review of fifty-six studies of fetal abnormalities found that they were not reliable enough indicators of trisomy for clinical practice.
• More than 95 percent of breast cancers have nothing to do with the BRCA genes.
• Less than half of those inheriting C282Y from both parents develop the biochemical abnormality of iron overload, and less than half of those with iron overload go on to develop the clinical disease.
• Fewer than two in a thousand people will be diagnosed with thyroid cancer in the next ten years.
• Absolute risk is a more important number to know than relative risk. The best estimate is that mammography reduces the death rate from breast cancer by about 20 percent. That’s a statement of relative risk. Here are the current absolute risks of breast cancer death for the typical fifty-year-old woman over the course of ten years: from about 5 per 1,000 without mammography to 4 per 1,000 with mammography. That means about 1,000 women have to be screened for ten years for one to benefit.
• Conditions with a net harm of diagnosis and treatment: borderline hypertension, borderline diabetes, small aortic aneurysm, small blood clots, screen-detected cancers, subtle fetal abnormalities, weakly penetrant genes
• Conditions with a net benefit for diagnosis and treatment: severe hypertension, severe diabetes, large aortic aneurysm, large blood clots, clinically detected cancers, obvious fetal abnormalities, highly penetrant genes
• False-positive results make up somewhere around 5 to 15 percent of all cancer screening results.
