4 ms·

It's worth remembering that a 36% increase of a marginal risk is still marginal. That number doesn't mean anything unless you know the original risk.

by Retra 11y ago

ekianjo 11y ago
Exactly. the typical risk of cancer is measured in "per million", so having a 36% risk does not mean "people are getting killed as we speak".
- Retra 11y ago
  At 35% per 10µg, it's actually about a 900% increase in lung cancer risk. That's a significant effect, because it means 900% more people will get lung cancer on average. People are getting killed, it's just hard to say how many without better numbers.
  - ekianjo 11y ago
    > At 35% per 10µg, it's actually about a 900% increase in lung cancer risk. You can extrapolate linearly such numbers, it makes absolutely no sense. This is not a math class, we are talking about clinical effects with actual thresholds.
    - Retra 11y ago
      "35% per 10µg" expresses a linear relationship. If someone says that's what it is, I'm going to assume they measured a linear relationship. It makes perfect sense if that is indeed the reality, and I see no reason why it cannot be. Unless you've got some actual conflicting data, it just sounds like you have unfavorable agenda.
      - ekianjo 11y ago
        Even if it's a linear relationship, it can only exist within the demonstrated range. You should know that not everything is linear when you go outside of the range you have studied. So unless proven this linear relationship is valid across all the spectrum (which is obviously is not), then you should take whatever is written with a big grain of salt and thing twice about it before making over simplistic assumptions.
- mikeyouse 11y ago
  Lung cancer has a normal occurrence of something like 60 in 100,000. If we are as conservative as we can be and say the increase is only 36% from all of these smog days, that would mean that the rate should increase to 82/100,000. Beijing's population is ~21 million. The 22 / 100,000 increase from the base rate would mean an additional 5,000 lung cancer deaths in Beijing due to the smog. That's the minimum. The more realistic number is surely an order of magnitude or more higher -- this level of pollution is unprecedented in modern times.
  - ekianjo 11y ago
    Again, what is the current data showing Beijing is far worse than the national average for lung cancer (because pollution has been pretty bad there for a long time already) ? I see assumptions, estimations but no hard facts to support anything, right now.
    - mikeyouse 11y ago
      I mean, you could try looking literally anywhere.. This is a well known problem; From 2014: http://www.bloomberg.com/bw/articles/2014-02-28/rates-of-lung-cancer-rising-steeply-in-smoggy-beijing http://www.bloomberg.com/bw/articles/2014-02-28/rates-of-lun... In 2002, for every 100,000 men living in Beijing, 49 had lung cancer. By 2010, that number had risen more than 50 percent, to 75. Women in China are less likely to smoke and so have lower rates of the disease. But among those living in Beijing, a similar sharp rise in lung cancer cases occurred over the last decade. In 2002, for every 100,000 women living in Beijing, 30 had lung cancer; by 2010, that number had also risen more than 50 percent, to 46. So the rate in the US is 60/100,000 -- As of 2010, in Beijing, the rate is closer to 121/100,000. Does a 100% increase in risk change your opinion of these 'marginal' deaths? For a population of 21M, this is equivalent to ~13,000 additional lung cancer deaths.
      - ekianjo 11y ago
        Problem is, lung cancer rates are increasing all around the globe, so that does not answer anything regarding "are the lung cancer rate in Beijing unusual vs the rest of the world?". But at least that's a start for a proper discussion. And please do not compare a US rate that includes countryside and cities with a city-based rate in China. That's not a relevant comparison at all. Rates vary wildy state by state in the US : http://www.cdc.gov/cancer/lung/statistics/state.htm http://www.cdc.gov/cancer/lung/statistics/state.htm As you can see the average in some states go to 90+ in the US.

It's worth remembering that a 36% increase of a marginal risk is still marginal. That number doesn't mean anything unless you know the original risk.

by Retra 11y ago

It's worth remembering that a 36% increase of a marginal risk is still marginal. That number doesn't mean anything unless you know the original risk.

ekianjo 11y ago

Exactly. the typical risk of cancer is measured in "per million", so having a 36% risk does not mean "people are getting killed as we speak".

Retra 11y ago
At 35% per 10µg, it's actually about a 900% increase in lung cancer risk. That's a significant effect, because it means 900% more people will get lung cancer on average. People are getting killed, it's just hard to say how many without better numbers.
- ekianjo 11y ago
  > At 35% per 10µg, it's actually about a 900% increase in lung cancer risk. You can extrapolate linearly such numbers, it makes absolutely no sense. This is not a math class, we are talking about clinical effects with actual thresholds.
  - Retra 11y ago
    "35% per 10µg" expresses a linear relationship. If someone says that's what it is, I'm going to assume they measured a linear relationship. It makes perfect sense if that is indeed the reality, and I see no reason why it cannot be. Unless you've got some actual conflicting data, it just sounds like you have unfavorable agenda.
    - ekianjo 11y ago
      Even if it's a linear relationship, it can only exist within the demonstrated range. You should know that not everything is linear when you go outside of the range you have studied. So unless proven this linear relationship is valid across all the spectrum (which is obviously is not), then you should take whatever is written with a big grain of salt and thing twice about it before making over simplistic assumptions.
mikeyouse 11y ago
Lung cancer has a normal occurrence of something like 60 in 100,000. If we are as conservative as we can be and say the increase is only 36% from all of these smog days, that would mean that the rate should increase to 82/100,000. Beijing's population is ~21 million. The 22 / 100,000 increase from the base rate would mean an additional 5,000 lung cancer deaths in Beijing due to the smog. That's the minimum. The more realistic number is surely an order of magnitude or more higher -- this level of pollution is unprecedented in modern times.
- ekianjo 11y ago
  Again, what is the current data showing Beijing is far worse than the national average for lung cancer (because pollution has been pretty bad there for a long time already) ? I see assumptions, estimations but no hard facts to support anything, right now.
  - mikeyouse 11y ago
    I mean, you could try looking literally anywhere.. This is a well known problem; From 2014: http://www.bloomberg.com/bw/articles/2014-02-28/rates-of-lung-cancer-rising-steeply-in-smoggy-beijing http://www.bloomberg.com/bw/articles/2014-02-28/rates-of-lun... In 2002, for every 100,000 men living in Beijing, 49 had lung cancer. By 2010, that number had risen more than 50 percent, to 75. Women in China are less likely to smoke and so have lower rates of the disease. But among those living in Beijing, a similar sharp rise in lung cancer cases occurred over the last decade. In 2002, for every 100,000 women living in Beijing, 30 had lung cancer; by 2010, that number had also risen more than 50 percent, to 46. So the rate in the US is 60/100,000 -- As of 2010, in Beijing, the rate is closer to 121/100,000. Does a 100% increase in risk change your opinion of these 'marginal' deaths? For a population of 21M, this is equivalent to ~13,000 additional lung cancer deaths.
    - ekianjo 11y ago
      Problem is, lung cancer rates are increasing all around the globe, so that does not answer anything regarding "are the lung cancer rate in Beijing unusual vs the rest of the world?". But at least that's a start for a proper discussion. And please do not compare a US rate that includes countryside and cities with a city-based rate in China. That's not a relevant comparison at all. Rates vary wildy state by state in the US : http://www.cdc.gov/cancer/lung/statistics/state.htm http://www.cdc.gov/cancer/lung/statistics/state.htm As you can see the average in some states go to 90+ in the US.