Search This Blog

Follow by Email

Friday, December 11, 2009

questioning everything: CT scans and cancer, coffee and diabetes

Computerized tomography, otherwise known as CT or Cat scanning, has imaged 70% of non-elderly adults in the last 3 years. The use of this technology has been steadily rising, and we now perform a total of over 19,500 CT scans per day in the US.

CT scans use a computer to organize x-ray data in such a way as to produce pictures that resemble cross sections of the human body, complete with bones, brains and soft tissues, tumors and blood vessels. The pictures are truly marvelous and have revolutionized the way we diagnose disease, allowing us to know many things about the insides of a person without actually cutting them open. We can see if a tumor is present, has spread, if an aneurysm is bursting or if the excruciating pain in a person's belly and back is a kidney stone or pancreatitis. We can tell if a victim of trauma is bleeding internally or if a mysterious fever is caused by a well hidden abscess.

In December's issue of the Archives of Internal Medicine, researchers from the National Cancer Institute and other participating institutions published an article looking at the cancer implications of all of these wonderful CT scans we have been doing. Based on risks of cancer predicted by a study of ionizing radiation done by the National Research Council, the CT scans done in the US in the year 2007 will be responsible for 29,000 cases of cancer, and 15,000 excess deaths. One CT scan of the abdomen and pelvis may be equivalent to 450 or more chest x-rays in terms of radiation exposure (though this varies by procedure and institution.)

CT scans, besides being life saving and revolutionary, are also killing us and eating up our health care budget. The trick is moderation. It is clear to me, from seeing my patients return from visits to emergency rooms or specialists, that we do far more CT scans than are truly necessary to diagnose serious disease. Sometimes an elective CT scan is interesting or reassuring, but just as often the tests done for interest or reassurance end up being confusing and anxiety provoking, as they show the benignly quirky internal makeup of individuals who might have cysts or enlargements of organs, duplicated spleens, liver hemangiomas, missing kidneys.

But moderation is not always a good thing. Take coffee, for instance. Another study reported in the month's Archives of Internal Medicine shows that a person who drinks coffee in large amounts has a significantly lower risk of getting diabetes than a person who drinks coffee moderately or not at all. Many studies have shown this, to varying degrees. The meta-analysis, a statistical combination of many small studies to produce a more robust result, suggests that for every additional cup of coffee you drink in a day, you reduce your risk of diabetes by 7%. Although fewer studies have looked at tea and decaf, they appear to carry the same benefits as real hi-test java. The best outcomes were seen in people who drank at least 6 cups of coffee a day.

Once again we see the limitations of science to address the concerns of the individual: if I drank that much coffee I would certainly die.

Tuesday, December 8, 2009

mammogram screening, take 2

I finally read through the 3 articles in the Annals of Internal Medicine that addressed the new recommendations from the US Preventive Services Task Force (USPSTF) about the recommendations for mammogram screening and breast exams. These have led to angry reactions, mostly based on lack of information and lack of understanding of the science behind the recommendations.

First of all, the USPSTF is far from the only organization to weigh in on screening recommendations. There are organizations such as the American Cancer Society, the American College of Obstetrics and Gynecology and various other official groups from various branches of medicine. The USPSTF is, however, the most evidence based of the groups, the least financially motivated, and the most conservative.

The recommendations of the USPSTF are categorized according to how sure they are that they are right, and changes come after long discussion and detailed evaluation of the research and the opinions of other organizations. In 1996 they were unable to endorse regular mammogram screening for women younger than age 50, but in 2002, based on information from studies done since that time, they extended their recommendations to women starting at age 40. They acknowledged at that time that there were risks associated with mammogram screening and that it was neither sensitive nor specific during that decade.

Yet more studies have become available since that time and the recommendations have gone back to encouraging women to begin mammogram screening at age 50, saying that screening before that time should “take patient context into account, including the patient’s values regarding specific benefits and harms.” They conclude that evidence is insufficient to recommend screening of women 75 years and older, which is solidly in line with recommendations in European countries. They do not, however, recommend against screening in older women.

The decision not to recommend mammograms routinely for women ages 40-49 is based on the fact that mammograms often pick up abnormalities of the breasts which appear suspicious, result in further procedures, and harm women through over treatment and excessive diagnostic testing. Breast cancers are detected in that decade, but the risks of screening all of those unaffected women outweigh the benefits of early detection in the few.

The recommendation that women not be taught breast self examination has engendered the greatest amount of misunderstanding of all. On the face of it, it makes no sense. What harm could self examinations do? Why should a woman not know her own body?

The recommendation is based on two large studies, one in Russia and another in China, evaluating self breast exams in a population that did not get regular mammograms. Women who were taught breast self examination techniques had no benefits in terms of breast cancer survival when compared to women who were not. So this recommendation really addresses the question of whether a physician should spend time specifically teaching women techniques of self breast examination. It does NOT say that doctors should now tell women not to examine their breasts. Perhaps women are just fine at examining their breasts without being harangued by their doctors to do so.

What, you may ask, is the rationale behind getting mammograms every 2 rather than every one year? Studies have shown that as many as 99% of breast cancers are picked up by every other year mammograms, and given known harms of radiation and associated costs of those extra mammograms, were they done every year, 2 year intervals seem like they are the magic number.

But why 2 years? What’s so special about the number 2?

At the Norwegian Institute of Public Health, studies have looked at women who got mammograms every two years for 6 years vs women who got only one mammogram at the end of 6 years. The women who got the mammograms every 2 years had a significantly higher incidence of breast cancer than the ones who got only one mammogram, and this difference persisted in the years that followed. The conclusion that the Norwegian doctors came up with was that the every 2 year screened women had breast cancers discovered which would have gone away if left untreated. American researchers have hotly disputed this interpretation, but their arguments are not compelling.

It is clear to me that the recommendations of the USPSTF to reduce recommended mammogram screening is NOT based on a politically motivated desire to reduce health care spending at the expense of the health of women. It is not yet clear to me what the best recommendations for mammogram screening should be. I intend to continue, as I have, to discuss with patients the pros and cons of mammogram screening and help them make the choices that are best in line with their risks and their values.

Monday, December 7, 2009

missing the safety net

What if you graduated from high school, left home, got a job delivering pizza, and were critically injured in a motor vehicle accident?

What if you had a part time job at a big company, a house, a family and got cancer?

What if you lost your job and your 8 year old daughter got appendicitis?

In the United States there are systems that act as safety nets for situations such as these, but they are not self sufficient and are severely strained in their ability to provide services with the progressive loss in adequate insurance coverage, the floundering economy and the increasingly outrageous costs of various forms of medical care.

If you were the first guy, ejected from your Geo Metro when you were t-boned at an intersection by a drunk driver, you would be taken to an emergency room at any hospital, transported to a trauma center if necessary, and treated until you were on the mend by that hospital. If you were eligible for medicaid or medicare due to the severity of your disability the hospital would eventually be reimbursed for the cost of your care (at least partly), and if you were not eligible, the hospital would attempt to bill you and when you were unable to pay, would eat the cost, part of which would be tax deductible.

If you were the second guy, the doctors who treated you would do so with little hope of being paid, might bill you, and would eventually eat the costs. You might be able to apply for a county emergency payment program to pay for things like surgery and CT scans, but you would eventually be expected to repay these costs. You would apply for public assistance based on disability, but the process of being approved for it would be slow. Your savings would inevitably be used up. You might lose your house.

In the case of the child, we have guaranteed medical insurance available for children through the government, but you do need to apply for it. The child with appendicitis might die or have some other bad outcome due to delay in treatment from lack of insurance. In a perceived emergency, though, treatment through the local emergency room would be assured.

Much of the problem with American health care stems from the escalating costs associated with it. Yet many of these costs, especially those associated with procedures and tests, go to hospitals who are the basis of our safety nets. In cutting costs, it is going to be vital that we pay attention to making sure that hospitals stay solvent. Providing adequate universal insurance will be a project that takes time, probably years. It is this insurance that can support hospitals and allow them to continue to support the communities in which they operate.

An article in the New England Journal of Medicine ( addressed our safety net system, and just how fragile it is. Because some hospitals are located in areas of particularly acute economic and social disaster, they are simultaneously vitally important to a safety net and totally inadequately reimbursed. Allowing hospitals like that to go under threatens the whole fabric of the larger medical system.

Saturday, December 5, 2009

How the conference on affordable health care went

There were nearly 30 people there, at a not much more than 40 bed hospital. This is unprecedented for a not-required noon meeting. There was food, but it wasn’t very good, so they were there for the content (or maybe they though the food would be better.)

The radiologist talked about appropriate use of technology and reducing unnecessary testing. Primary care docs talked about ways to make the computerized medical record systems give information about costs. The pharmacist talked about how to find out good information about drug costs, and we discussed ways to educate docs in the hospital on alternatives to the most expensive medications. We discussed other methods for reducing pharmacy costs which will also have other health benefits (changing medications given by vein to ones given by mouth, for instance.) The hospital CEO was willing to commit to putting into practice a system that would promote cost transparency for providers and patients. We talked about shifting responsibilities for record keeping to nurses in our offices so we would have more time to see patients so they wouldn’t have to go to emergency rooms where the costs are higher and the care is less personalized.

What will come of this is unclear, but the very act of discussing it in an open forum is brand new in my experience.

There is a part in the hippocratic oath about sharing the precepts and learning only with those who have taken the sacred oaths and the sons of other doctors. This may partially underlie a tendency of doctors to be circumspect. For whatever reason the workings of the practice of medicine are not shared easily outside of the profession. Discussing and re-evaluating what we do will go against some pretty basic instincts.

Post conference feedback has been interesting too. As might be imagined, not everybody had their say, and not everybody's issue got discussed. One provider mentioned that she would have liked to problem solve some really pressing issues of access to care. A tech guy mentioned that he had lots of ideas on how to make the hospital staff more efficient by making technology more effective. I am sure that the more we talk, the more issues will come up, and there may need to be smaller conversations and groups of people with similar interests will have to do their own work.