Do we talk about the right cancers?

By John Pickering 12/12/2012 7

We all want a cure for cancer.  This is evident in the ideas put forward for the NZ Science Challenges (  Will these ideas reflect the true need in New Zealand? Are all cancers equal or are some more equal than others?  Does the public perception of what is most important reflect the true health risks?  To answer this latter question would require a well controlled survey which I don’t have.  What I can do is look at some data from the media.

First the numbers: The Table below gives the number of newly diagnosed registered and of deaths in New Zealand for various cancers for 2009 (I have only considered those with more than 200 deaths).  The percentages are the percentage of all the cancers listed.  The final column is the ratio of the numbers who died to the number of newly diagnosed.  This is not a proper measure of the likelihood of dying of the disease once it is diagnosed, but probably gives an impression of which are the more and which are the less deadly of the cancers.

Cancer Registrations and Deaths in New Zealand (2009: Ministry of Health,

Cancer Registrations and Deaths in New Zealand (2009: Ministry of Health,

The graph plots the percentages for new Diagnoses and deaths against the percentage of stories on the web site for each of the cancers (ie relative to the total number of stories (20,757) for all these cancers.  I’m sorry to pick on Stuff, but it conveniently gave a count with a fast search engine.  The area above the dashed line (blue) suggests under reporting in the media, the area below over reporting.  Brain cancer is vastly over reported on this measure relative to other cancer death rates and new registrations.  Breast cancer is also over reported, particularly in relation to other cancer death rates.  Melanoma reporting reflects its deadliness, but not its new registrations.  Colorectal/Bowel cancer and Lung cancer are relatively poorly discussed compared to their relative death rates.

The percentage of new registrations and of deaths of various cancers against the percentage of stories in  Note, percentages are relative to the total number of new registrations/deaths/stories

The percentage of new registrations and of deaths of various cancers against the percentage of stories in Note, percentages are relative to the total number of new registrations/deaths/stories

There may well be many better ways to look at this kind of data – I would appreciate any suggestions.  I hope to look at other diseases as well.  However, as I research a disease that affects as many people as breast cancer each year, is just as deadly, and yet is reported on Stuff 1/50th as often, I am a little biased.

Tagged: brain cancer, breast cancer, cancer, lung cancer, media, melanoma

7 Responses to “Do we talk about the right cancers?”

  • I am not sure where you coming from or where you’re going with this. In particular, I struggle with the word “equal”. You do probably know that not all cancers are similar. In fact, inter- and intra-patient but also intra-tumour heterogeneity is enormous; one tumour can contain very many mutations, for example, but these will not be evenly spread throughout the tumour. The cancers in the table are listed by site but there are a number of breast cancers, for example; triple-negative breast cancer is known to spread to the brain so is this a brain or a breast cancer and what is the cause of death? The other issue is ease and time of diagnosis; early diagnosis can save lives. The Ratio in the table is a little problematic. For example, in 2009 there were 6 registrations of cancer of the heart (C38) and 9 deaths, which gives a ratio of 1.50. The point is that from the numbers in the table alone you cannot draw conclusions about prognosis and life expectancy. In total there were 8437 deaths from all cancers in 2009 and 20875 registrations, which gives an average ratio of 0.40 (0.43 in 2007 and 0.42 in 2008).

    You also mentioned “public perception”. That is a hot potato. A very recent study reported in the New England Journal of Medicine was entitled “Patients’ Expectations about Effects of Chemotherapy for Advanced Cancer” (Preview is free; and concluded that “Many patients receiving chemotherapy for incurable cancers may not understand that chemotherapy is unlikely to be curative” and went on to say “Physicians may be able to improve patients’ understanding, but this may come at the cost of patients’ satisfaction with them”. I think it is fair to say that in NZ there is similar levels of confusion and lack of understanding, certainly among the wider general public. Cancer is enormously complex and it is therefore extremely difficult to develop successful therapies.

  • Thanks Frederick. “equal” comes from a famous quote from Animal Farm by Geore Orwell – “all animals are equal, but some animals are more equal than others.” Sorry for being so obtuse.
    I indicated the ratio was not a great measure…as you show it fails with small numbers (which I purposely avoided), it also fails if there have been differences in changes in rates of diagnosis and death.
    Very interesting comment about perception. I think this is important. Physicians are caught between a rock and a hard place. Important that there is more public education.

  • Great post John! Fantastic and (relatively) quick analysis of easily accessed data which tells a good story. What do you think drives the media fascination with brain cancer?

  • I’m not sure that there is a clear point, but I do find this an interesting subject to explore. The graph is fascinating and I wonder what similar from other news agencies would be like.

    Of the top five cancers in incidence, lung, bowel, skin and prostate all appear under-reported, whereas breast appears relatively over-reported.

    Is this just a function of Stuff’s reporting? Or could it be that it is representative of all media? Why should there be a difference anyway?

    Brain cancer appears to be vastly over-reported by Stuff, but that’s an impression I’ve got from all media. My impression is that brain cancer reporting is often along the lines of “X…with brain cancer…incurable by conventional medicine…seeking help overseas…unconventional treatment…only available from…expensive.” Does anyone else get that impression of brain cancer reporting?

  • It’d be easy to over interpret with this data. I looked at the Herald site and found:
    “Brain cancer” 65
    Breast cancer 1000 (exactly? Is this a max for the site?)
    Prostate cancer 509
    Lung cancer 522
    This suggests to me that it is a “Stuff” thing – I’m a bit suspicious of the way articles are counted.

    Skimming articles – there seem many that are celeb driven – either as a celeb getting a particular cancer or helping someone who has one. So if celeb A gets cancer of the Big Toe, expect it to be over represented. There is a positive in that it can highlight a previous rare event.

    I’m not convinced I’ve got things right here. I may have a play and try again next week with a few other diseases.

    What would be really fascinating would be to what extent donations to disease specific charities reflect media exposure or disease incidence or disease risk. Anyone with a spare student looking for an interesting project?

  • You may have seen the article in Saturday’s NZ Herald “People living longer but sicker: study” The reference is to a Feature in The Lancet “Global Burden of Disease Study 2010” I have not read it yet (maybe in the holidays) but it shows that limiting searches to Stuff, NZ Herald, or fails to acknowledge that NZ is but a very small country in large world in which information is (almost) freely available. It also ignores that cancer research in NZ is not just done for NZ patients. In fact, when it comes to finding a commercial ‘vehicle’ to fund oncology trials one inevitably ends up somewhere outside NZ, both in investors and in patients. Having said that, NZ cancer researchers often manage to find clinicians and ‘clinical sites’ in NZ to participate in those trials. There a great many oncology trials conducted in NZ.

Site Meter