Archive | Epidemiology RSS for this section

Bad Medicine: Spycraft and Vaccination Efforts

120px-CIA.svgI admire many of the creative efforts used by our intelligence agencies to keep America safe.  Spycraft itself is a wonderfully fascinating area, both in reality and fiction.  But some areas should be off-limits.

The 2011 intelligence effort to locate Osama bin Laden included a vaccination program aimed at obtaining DNA from children at a compound where bin Laden was suspected of – and later found to be – hiding.

A Pakistani doctor involved with the effort remains in legal trouble for his involvement in the plot, and local suspicion of future vaccination efforts in the aftermath of this scheme has led to an increase in polio cases.  Fortunately, the White House has agreed to cease any future use of vaccination programs as cover for intelligence operations.

Vaccination has been one of medicine’s great triumphs.  Leaving this area off-limits for intelligence gives assurance to the world that vaccination programs can always be viewed as legitimate.

Polio Vaccination, India 2002

This will keep polio out of your future

The anti-vaccination effort in this country has been an unfortunate development, sowing doubt via unscientific and outright fraudulent information.  No further seeds of doubt should be developed, and it’s a good thing that the CDC continues to campaign against such misinformation.  By putting vaccination programs off-limits, the CIA has taken an important step and owned-up to its past.

Creativity from our intelligence services is normally a good thing.  Reality and fiction have given us everything from exploding cigars to sports-car submarines.  At least they will no longer give us vaccination ruses.

I’m glad we got bin Laden.  I hope we did not truly need the vaccination program to do so.


No Good Link Between Cell Phones and Brain Cancer

Recent reports have surfaced noting that cell phones appear to cause increased brain activity due to the non-ionizing radiation that they emit.  This has rekindled the debate of whether or not cell phones are a causative factor for primary brain cancer.  The official word on this can be found here, and suffice it to say that so far there has been no definitive proof.

I wanted to take a 30,000 foot view of the problem, by comparing primary brain cancer rates with cell phone usage in the United States.  Note that “primary” means a central nervous system tumor that is not a metastasis from another site.

Below is a composite graph I created demonstrating the relationship. (Click for larger version)

It is a little confusing but I will explain.  The blue line indicates the takeoff of cell phone subscribers in the U.S. over time.  The scale for this line is on the left of the diagram.  It reaches upwards of 200 million by 2007.  The red sawtooth line is the incidence of primary brain cancers in the U.S. for the period 1975-2007.   The scale for this line is on the right, and it shows that the incidence has varied between 5.8 and 7.  This is the number of new cases of brain/central nervous system cancers per 100,000 people, and is age-adjusted.

Naturally, this question requires more detailed epidemiological analysis, but the early indication is that there is no obvious increase in brain cancer during the period that cell phone use became widespread.  Rates could increase in the future if there is tissue damage that occurs over time, but to date that does not seem to be the case.

Like most other expert opinions in medicine, links such as “cell phone use and brain cancer” have to be taken with a grain of salt.  The link between lung cancer and smoking was an epidemiologic triumph, but since that time, many other links have been discredited.  Consider these:

Pancreatic cancer and coffee (No)
Hot dogs and childhood leukemia (No)
Vaccines and autism (No)
“Western diet” and cardiovascular disease (No)
Silicone breast implants and autoimmune disease (No)

For more, check out the book:   The Rise and Fall of Modern Medicine.

But before real science discredits the link, public impressions can be formed and can be very difficult to reverse through education.  Often, representing the question with a simple graph can be easier to understand.

National Cancer Institute (SEER)
Cell phone subscriber data

Subscriber data for 2007 was not available, so was estimated based on 2006/2008 data.
Cancer incidence from 2008-2010 not available, but data from the American Cancer Society estimate new cases of CNS malignancies at 22,000 for each of these three years, remaining stable when cell phone use has only increased.

Spiral CT Scans For Lung Cancer Screening: High Cost, Modest Benefit

Beschreibung: Kleiner Lungentumor (T1) linker ...

Image via Wikipedia

A new study has shown that  spiral CT scan screening of heavy smokers can reduce deaths from lung cancer.  The study looked at 53,000 heavy smokers aged 55-74.  Patients were screened with either chest x-rays or spiral CT scans.  There were 354 deaths in the CT group vs. 442 deaths in the x-ray group over an eight-year period.  The difference was significant, but it is not clear whether spiral CT scans should be used as a standard-of-care screening test in this population.  Concerns such as radiation risk and false positives warrant further exploration, but for now I will look at the costs.

Sources put the cost of a CT scan between $300 and $400.  Let’s split the difference and call it $350.  The study was randomized, so assume that 26,500 people had CT scans at a cost of $9,275,000.  This resulted in 88 fewer deaths, or a preventive cost of $105,397.72 each.  Does this number have meaning?  A common cost/benefit reference for costly treatments, such as cancer therapy, is $50,000 per year of life gained.  It seems likely that this test would be cost effective by this standard, since odds are good that those 88 individuals should live another 2 yrs each, many quite a few more.

The study probably represents an incremental benefit against lung cancer.  The scan does nothing to the biology of the disease, but it increases the odds of successful treatment with earlier detection.

The statistics can be looked at from another angle, one which makes the CT test seem less attractive.   To prevent 88 deaths, 26,500 people were screened, or about 300 scans per single death prevented.  Preventing 20% of the 159,000 yearly deaths from lung cancer would save 31,800 lives.  But, it would require CT scans of 9.5 million people at an annual cost of more than $3 billion.

To be fair, my analysis does not take into account the cost implications of preventing these deaths, nor does it address the costs of chasing false positive scans or several other “what if” scenarios.

These types of analyses will become necessary for many types of treatments and tests as we convert healthcare to a much more evidence-based system than we currently have.

Lung cancer is a nasty disease with nasty outcomes, and smoking is a very hard habit to break.  The question of how much to spend on smoking cessation efforts and how much to spend on detection and treatment of lung cancer is not an easy one.

The National Lung Screening Trial