4 ms·

You also expect the number of identified cases to increase as we improve detection methods and find tumors earlier.

by eli 9y ago

FLUX-YOU 9y ago
53 million is also not necessarily a population 100% seeking healthcare either, so if more people are getting access healthcare, more tumors will be found.
- pacaro 9y ago
  While it’s not perfect, it’s a reasonable assumption in the U.K. that everyone has access to healthcare. TBH it’s a reasonable assumption in every developed country except one.
  - krageon 9y ago
    I would posit that if not everyone has access to healthcare, it's not a developed country.
  - ghufran_syed 9y ago
    Except access to a doctor is not the same as access to the key test, usually a CT in this case. In the NHS, the general practitioners are explicitly set up to be “gatekeepers” for expensive testing or referral to specialists. So it’s not hard to imagine older patients with symptoms having ct’s delayed in the past for some period of time - and some proportion of these patients, usually older patients, will die of something else before their GBM is discovered. But as people die less of other things, it could uncover more gbm’s Having worked in the UK and in the US health systems, in general in the UK you have more of a risk of something that could be done, not being done (to save the government money). Meanwhile, in the US, if you have good insurance, you are more at risk of over-testing and over-treatment (because it reduces the doctor’s perceived legal and complaint risk, and sometimes makes the doctor or hospital more money)
gnode 9y ago
Is this actually likely to have a notable impact on the statistics? My understanding is that cases diagnosed later or discovered in autopsy will contribute just as much as cases diagnosed early. Assuming that such tumours aren't routinely missed in autopsies. I imagine improvements in early detection may result in a temporary increase due to the shifting phase of detection, but this wouldn't account for a doubling over twenty years.
- astura 9y ago
  1) Most people aren't autopsied upon death, autopsies are only performed in limited circumstances. I don't personally know of a single person whom died who an autopsy was performed on. 2) The article is vague and short but "cases of GBM in England have increased from around 1,250 a year in 1995 to just under 3,000" almost certainly refers to diagnosed cases of GBM, not including incidentals found in autopsies.
  - gnode 9y ago
    Autopsies are performed when the cause of death is indeterminate. Most people die of a known cause. A family member of mine was autopsied (in the UK) when they died unexpectedly. I expect that if killed by brain cancer, you'll either present symptoms resulting in a diagnosis (albeit a late one), or die unexpectedly and be autopsied.
    - astura 9y ago
      You're entirely ignoring comorbidity. Glioblastomas are associated with Alzheimer's disease. Its entirely possible that 20 years ago the symptoms of a brain tumor would be attributed to Alzheimer's when the patient was already dying of Alzheimer's, or other terminal illness for that matter. With better awareness and detection we may pick up new diagnoses in the sick and dying.
carbocation 9y ago
> You also expect the number of identified cases to increase as we improve detection methods and find tumors earlier. If this is true (and I think you're likely to be right) then we should see longer survival with GBM thanks to lead-time bias.
- awakeasleep 9y ago
  It's actually more complicated than that- we'd need to know the survival rate for procedures carried out at different discovery times. If the cancer is basically inoperable, or if there isn't a big difference in outcome based on time of detection, then there wouldn't be an appreciable change in survival rates
  - carbocation 9y ago
    > If the cancer is basically inoperable, or if there isn't a big difference in outcome based on time of detection, then there wouldn't be an appreciable change in survival rates Lead time bias tells you that even if there is no effective treatment, you will appear to live longer thanks to earlier discovery of the disease. It was a hot topic with regard to prostate cancer a few years ago.
brudgers 9y ago
Going further, I'd expect the number of unidentified/misidentified cases to decrease as information flows more freely alongside the internet's growth since 1995. Today, a doctor can investigate a brain tumor from their phone in ways that were impossible using a top research library in 1995. In particular, the internet is really good at surfacing long tail information, and a rare cancer is by definition out in the long tail. Today, I can go from about zero knowledge about the subject to a remarkable level of awareness the within seconds deciding to add an example. https://en.wikipedia.org/wiki/Glioma https://en.wikipedia.org/wiki/Glioma The world today is such that I get that awareness increase while I am typing this comment. The hammer makes me aware of nails.

You also expect the number of identified cases to increase as we improve detection methods and find tumors earlier.

by eli 9y ago

You also expect the number of identified cases to increase as we improve detection methods and find tumors earlier.

53 million is also not necessarily a population 100% seeking healthcare either, so if more people are getting access healthcare, more tumors will be found.

pacaro 9y ago
While it’s not perfect, it’s a reasonable assumption in the U.K. that everyone has access to healthcare. TBH it’s a reasonable assumption in every developed country except one.
- krageon 9y ago
  I would posit that if not everyone has access to healthcare, it's not a developed country.
- ghufran_syed 9y ago
  Except access to a doctor is not the same as access to the key test, usually a CT in this case. In the NHS, the general practitioners are explicitly set up to be “gatekeepers” for expensive testing or referral to specialists. So it’s not hard to imagine older patients with symptoms having ct’s delayed in the past for some period of time - and some proportion of these patients, usually older patients, will die of something else before their GBM is discovered. But as people die less of other things, it could uncover more gbm’s Having worked in the UK and in the US health systems, in general in the UK you have more of a risk of something that could be done, not being done (to save the government money). Meanwhile, in the US, if you have good insurance, you are more at risk of over-testing and over-treatment (because it reduces the doctor’s perceived legal and complaint risk, and sometimes makes the doctor or hospital more money)

gnode 9y ago

Is this actually likely to have a notable impact on the statistics? My understanding is that cases diagnosed later or discovered in autopsy will contribute just as much as cases diagnosed early. Assuming that such tumours aren't routinely missed in autopsies. I imagine improvements in early detection may result in a temporary increase due to the shifting phase of detection, but this wouldn't account for a doubling over twenty years.

astura 9y ago
1) Most people aren't autopsied upon death, autopsies are only performed in limited circumstances. I don't personally know of a single person whom died who an autopsy was performed on. 2) The article is vague and short but "cases of GBM in England have increased from around 1,250 a year in 1995 to just under 3,000" almost certainly refers to diagnosed cases of GBM, not including incidentals found in autopsies.
- gnode 9y ago
  Autopsies are performed when the cause of death is indeterminate. Most people die of a known cause. A family member of mine was autopsied (in the UK) when they died unexpectedly. I expect that if killed by brain cancer, you'll either present symptoms resulting in a diagnosis (albeit a late one), or die unexpectedly and be autopsied.
  - astura 9y ago
    You're entirely ignoring comorbidity. Glioblastomas are associated with Alzheimer's disease. Its entirely possible that 20 years ago the symptoms of a brain tumor would be attributed to Alzheimer's when the patient was already dying of Alzheimer's, or other terminal illness for that matter. With better awareness and detection we may pick up new diagnoses in the sick and dying.

carbocation 9y ago

> You also expect the number of identified cases to increase as we improve detection methods and find tumors earlier. If this is true (and I think you're likely to be right) then we should see longer survival with GBM thanks to lead-time bias.

awakeasleep 9y ago
It's actually more complicated than that- we'd need to know the survival rate for procedures carried out at different discovery times. If the cancer is basically inoperable, or if there isn't a big difference in outcome based on time of detection, then there wouldn't be an appreciable change in survival rates
- carbocation 9y ago
  > If the cancer is basically inoperable, or if there isn't a big difference in outcome based on time of detection, then there wouldn't be an appreciable change in survival rates Lead time bias tells you that even if there is no effective treatment, you will appear to live longer thanks to earlier discovery of the disease. It was a hot topic with regard to prostate cancer a few years ago.

brudgers 9y ago

Going further, I'd expect the number of unidentified/misidentified cases to decrease as information flows more freely alongside the internet's growth since 1995. Today, a doctor can investigate a brain tumor from their phone in ways that were impossible using a top research library in 1995. In particular, the internet is really good at surfacing long tail information, and a rare cancer is by definition out in the long tail. Today, I can go from about zero knowledge about the subject to a remarkable level of awareness the within seconds deciding to add an example. https://en.wikipedia.org/wiki/Glioma https://en.wikipedia.org/wiki/Glioma The world today is such that I get that awareness increase while I am typing this comment. The hammer makes me aware of nails.