What we can learn from cancer statistics

Consider a hypothetical world in which there are four equally possible causes of death – cancer, heart disease, infectious disease, and accidents. Now imagine that, through a miracle of medicine and public health we were to eliminate all deaths from infectious disease – what would happen to our causes of death? All else being equal, we’d expect to see mortality from cancer, heart disease, and accidents rise from 25% to 33% each – mortality rates from each of these would jump by a third.

Now think a little further – infectious disease is an equal opportunity killer that hits the young as well as old, while heart disease and cancer tend to strike later in life. Reducing deaths among younger people lets more of us live into our cancer- and heart-disease-prone years, so of course these will begin to take a greater toll. As public health and medicine (not to mention other safety measures that reduce accidental deaths) lengthen our lives we should expect to see more people succumbing to things like cancer and heart disease – and whenever we think about reasons for increases in these diseases of our later years we have to try to untangle the impact of longer lives from the impact of whatever possible we’re considering. This isn’t to say that our lifestyles, eating habits, and environment play no part in the rising numbers of death due to cancer and heart disease, just that we have to be able to disentangle the effects of a longer lifespan from the effect of whatever cause(s) we are examining.

This came to me as a sort of epiphany when I was in Cambodia several years ago  on a radiation safety mission – I was visiting Cambodia’s sole radiation oncology clinic. I asked the facility’s sole medical physicist why there was only one clinic, expecting him to talk about governmental instability or poverty. Instead he pointed out that the average Cambodian didn’t live long enough to get cancer – with a median lifespan of less than 50 years most Cambodians died of malnutrition, injuries from landmines, disease, and simple poverty. He went on to say that he’d consider it a triumph to have to open a second clinic because it would mean that more Cambodians were living long enough to get cancer.

Interestingly, in spite of huge increases in human exposure to synthetic chemicals, electromagnetic fields, cell phone radiation, food additives, ionizing radiation, and so forth, statistics maintained by the American government clearly show that the age-adjusted cancer incidence has been dropping fairly steadily for nearly a century. Of course, this doesn’t mean that none of these things causes cancer; simply that whatever impact they have (at the levels to which most of us are exposed) is likely lower than the gains we derive coupled with improved screening and treatment. As one example, the radiation from x-rays might be carcinogenic (although at these very low levels of radiation exposure it’s almost impossible to say one way or the other), but we also derive a huge amount of good from being able to properly diagnose injuries and disease with these x-rays. Thus, medical x-rays on the whole are likely to extend life more than to cut it short because they can catch things that are immediately dangerous while adding only very slightly (if at all) to the risk of developing cancer in another few decades.

We should also keep in mind that, even though the incidence of cancer is dropping, the total number of people who get cancer has been steadily rising. Part of the reason for this is that the population of the US keeps growing – even a slightly decreasing cancer rate can lead to more cancer cases if the population grows more rapidly than the cancer rate drops (for example: say that 30% of the people in a city of a million people develop cancer so that 300,000 people develop cancer in their lifetimes – if the cancer rate shrinks to 20% while the population grows to 2 million then there will be 400,000 people who get cancer). Paradoxically, a reduced cancer rate can still lead to more total cancers if the population size increases rapidly enough.

Something else to throw into the mix is adjusting for age – cancer is more likely to be an affliction of an older population so as a population ages it is also more likely to develop cancer (as I saw in Cambodia). So to get a fuller picture of what’s going on it’s only fair to compare the cancer rates in each age bracket – instead of looking at gross numbers, or even overall rates, of cancer incidence it’s better to compare people of similar age. So we look to see how cancer rates among, say, 80 year-olds changes from year to year, along with cancer rates among those in their 70s and so forth – this is the age-adjusted cancer rate. And as noted above, what we find is that the age-adjusted cancer incidence (with the exception of smoking-related cancers) has been dropping steadily for nearly a century.

What this all means is hard to say – I’m not an epidemiologist and I don’t want to jump to unwarranted conclusions. But it at least suggests that many of the things we are concerned about (electromagnetic fields, radiofrequency radiation, cell phones, the level of radiation found in medical x-rays, and so forth) – things to which our exposure has skyrocketed over the last several decades without a concomitant increase in cancer rates – might not be as bad as we fear. Speaking about the health and environmental impact of radiation and its variability radiation biologist Antone Brooks made a comment about ionizing radiation – that whatever the effect is at low levels of exposure “it’s not a big player.” When exposure to medical radiation and electromagnetic fields have increased by a factor of thousands while cancer rates have dropped it is reasonable to ask if our concerns might outstrip the data.

Tags: ,

4 Responses to “What we can learn from cancer statistics”

  1. Bob Applebaum March 1, 2012 at 10:28 AM #

    Just a bit of fine-tuning…the incidence (new cases per time, usually “per year”)of cancer has been generally increasing over the last half century. It is the death rates which have been generally decreasing due to improved diagnostics and therapeutics.

    To discuss a statistic like “..300,000 people develop cancer in their lifetime…” is called lifetime prevalence and shouldn’t be confused with incidence.

    It also doesn’t really make sense to mention our “skyrocketing” exposure to things like EMF, cell phones, and RF radiation when those things are not known carcinogens.

    We know medical x-rays do increase the risk of cancer, though the risk increase is small.

    • Derek March 1, 2012 at 11:46 PM #

      Actually we do not know if medical x-rays increase the risk of cancer. The only thing we know is that there is an increase in cancer risk at or above 10,000 mrem of acute (all at once) whole body radiation exposure (it’s about a 1% increased risk at 10,000 mrem). Below 10,000 mrem there is no data to unequivocally support this conclusion BUT the regulations assume a linear relationship between cancer risk and exposure all the way down to zero. That’s why we say “there is a risk.”

      FYI – Medical x-ray doses range 0.5-800 mrem.

      • Ian March 6, 2012 at 5:03 AM #

        Small correction; we do not know that there is an increased risk of cancer above 10,000mrem. Instead, it is above that level that other non-cancerous acute affects can be observed. (burning, cell death etc)

        The Linear No Threshold model is wrong, and everyone knows it is. But as it is pessimistic it is fine to use for regulatory purposes (i.e. to calculate safe exposure limits).

    • Dr Y March 5, 2012 at 11:32 AM #

      Bob – you have some good points. The reason I mentioned the increased exposure to EMF, cell phones, etc. is because they have been cited by others as being potential carcinogens. I remember, for example, the big controversy over high-tension power lines in teh 1980s and today’s controversy over mobile phones and what has been called “electromagnetic smog.” I think that I was just not clear in the way that I presented this and I apologize.

      With regards to medical radiation, it is certain that our exposure to medical radiation has shot up dramatically in just the last few decades, although the majority of exposures fall below the 5-rem (50 mSv) level at which the Health Physics Society states it is inappropriate to calculate a risk estimate due to the fuzziness of the epidemiological data. For this reason I am not sure that I would agree with so broad a statement as you made – I’d feel more comfortable saying that “high levels of medical radiation exposure can increase our risk of cancer” – but I completely agree with you that whatever risk that exists is small, and most likely offset by the benefit from the diagnostic information x-rays provide.

      Many thanks for your comments – and for your fine-tuning!

Leave a Reply