Sunday, 19 August 2012

My chart vs CRUK chart


Source CRUK
The continuing rise of lung cancer in women reflects the high number of female smokers several decades ago when attitudes were different. Tobacco advertising hasn’t appeared on UK television since 1965, but that didn’t stop the marketing of cigarettes.- Jean King, director of tobacco control. (Cancer Research UK)

I have long been interested in this chart produced by Cancer Research UK for a number of reasons.

Firstly, I wonder why they do not show the rate of lung cancer before 1975 because the UK like many other countries have kept data from much further back in time. Mexico for example has collected data since the 1950s just like Scotland has. Sweden has excellent records going back even further. CRUK decided to use incidence data rather than mortality data and it maybe that incidence data only goes back to 1975 but I am sure that CRUK could have chosen to use mortality data if they had wanted to.


Secondly, It would seem to the casual observer that the decline in the rate of lung cancer for males could very well be caused by the decline in smoking prevelance. And that the rate for lung cancer could be at around the 100+ mark from 1948 through to 1975 for males. But as this is not shown it would be just guessing. However, the only thing that may seem to conflict with this theory is why there is no similar decline in lung cancer for females after 1975 when both sexes were either not taking up smoking or becoming quitters. I have heard it asked in the media on a number of occasions to "cancer experts" by people asking the very reasonable question "why is lung cancer still rising for women"? Every time the response goes like this "There has been a greater fall in the number of men smoking compared to women and this is why there is a rise for women" or along those lines. This is risible nonsense. The two are not connected. The reward for women not smoking is more lung cancer, therefore the theory that smoking protects women from lung cancer holds more water than the theory that smoking causes lung cancer in women. I would say the correct response to the question of why lung cancer is on the rise for women is "We don't know why lung cancer is still rising for women". There is nothing wrong in my book with saying "We don't know". So why so called "cancer experts" insist on trying to pin this rise in lung cancer for women on smoking is beyond me. What ever way you look at it, it is clear that women have been giving up smoking since 1975 just like men, the reward for men is less lung cancer but the reward for women is more lung cancer.

Thirdly, the smoking prevelance is data wieghted after 1998, again this is not CRUKs fault, if the data has been wieghted than they must be shown wieghted but it does mean two different types of measurements have been used on the same trend line.


I have long wanted to produce my own chart showing similar data to illustrate the point that the theory that smoking prevelence drives lung cancer becomes more confused when you look further back in time. My first idea was to produce a chart the same in all respects but with using mortality data going back to 1950. But I decided instead to use smoking prevelance for middle aged people aged 35 - 59 and lung cancer data for a thin slice of population ages 60 - 64 (the five years after the upper limit for smoking prevelance data). The trends are broadly speaking about the same as for a chart showing all ages.

Here it is.

Cancer stats source
The continuing rise of lung cancer in women reflects the high number of female smokers several decades ago when attitudes were different. Tobacco advertising hasn’t appeared on UK television since 1965, but that didn’t stop the marketing of cigarettes.- Jean King, director of tobacco control. (Cancer Research UK)
I simply don't understand how one can look at these trends and come to the conclusion that cigarette use drives lung cancer. How can a fall in smoking prevelance (males) up to 1975 cause a rise in lung cancer? At this time why were women not getting so much lung cancer and why did they get more of it later , why not earlier like the men? If it is true that smoking causes 95% of lung cancer then I would expect smoking prevelance and lung cancer rates to be highly correleated and to me they seem only weakly correlated. One correlation bieng that as there was a rise of lung cancer amd people gave up smoking or did not take it up in the first place. Had there not been a massive rise in lung cancer maybe more people would still be smoking.

As far as I can see cigarette consumption does not drive lung cancer rates in the UK. This does not mean that smokers are not more likely to get lung cancer it just means something else drives lung cancer rates.

Friday, 30 March 2012

What ever happened to the cigarette caused lung cancer epidemic in Mexico?



All numbers for the year 2000

SOURCE


It is astonishing to think that even if the entire population of Scotland gave up smoking tomorrow the chances are that their lung cancer death count would still be about the same as Mexico.
That the number of deaths of lung cancer for Mexican men aged 70-74 is only hundreds more than in Scotland is also astonishing. One might think that Mexicans have some genetic advantage but clearly this did not help the Spanish.