3 ms·

A couple of data points. It looks like 20% is only in women - in men, it's 9%. Second, the ratio of smokers:nonsmokers can be at least partially explained by t

A couple of data points. It looks like 20% is only in women - in men, it's 9%. Second, the ratio of smokers:nonsmokers can be at least partially explained by the number of smokers decreasing. Third, it's not clear whether this is a real change, or if it's a data anomaly. Personally, I would be very careful about the confidence with which I express this perspective. It might be true, but it's not at all clear that it is, so it's not clear there is a mystery. “It is well-documented that approximately 20% of lung cancer cases that occur in women in the U.S. and 9% of cases in men, are diagnosed in never-smokers,” "To a great extent, this is a function of straightforward math... Fewer people smoke today than in previous decades... because there are fewer smokers in the population, out of every 100 lung cancer patients, fewer will therefore be smokers. And that means more of them will be never-smokers." "There are also hints that the absolute incidence of lung cancer in never-smokers has been rising, said oncologist John Heymach of MD Anderson Cancer Center. Some data say it has, but other data say no. The stumbling block is that old datasets often don’t indicate a lung cancer patient’s smoking status, Heymach said, making it impossible to calculate what percent of never-smokers in past decades developed lung cancer." https://www.statnews.com/2021/01/26/growing-share-of-lung-cancer-turning-up-in-never-smokers/#:~:text=%E2%80%9CIt%20is%20well%2Ddocumented%20that,of%20the%20American%20Cancer%20Society https://www.statnews.com/2021/01/26/growing-share-of-lung-ca....

incomingpain 4y ago
I appreciate the response. I probably won't be able to reply, but I guess i'll be wasting my time typing a response. Sorry if I dont put as much effort in; sourcing etc. >A couple of data points. It looks like 20% is only in women - in men, it's 9%. Second, the ratio of smokers:nonsmokers can be at least partially explained by the number of smokers decreasing. Third, it's not clear whether this is a real change, or if it's a data anomaly. This is something I didn't fully understand and had to research. Why does it disproportionately affecting black people and women more? Nobody really explains it but the answer is known. Yes absolutely, as # of smokers decreases we would expect it to increase. Totally, but the correlation between decrease in tobacco usage and increase in lung cancer inverted. I think that's why when we saw an increase in 2006ish nobody responded like they have been responding in the last 5 or so years. >Personally, I would be very careful about the confidence with which I express this perspective. It might be true, but it's not at all clear that it is, so it's not clear there is a mystery. I hear your threat and yes I'm flagged and such. >“It is well-documented that approximately 20% of lung cancer cases that occur in women in the U.S. and 9% of cases in men, are diagnosed in never-smokers,” I believe what you are referencing is: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2764546/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2764546/ This study was done in 2006 publishing in 2007. Yes absolutely. 20% vs 9% was certainly a thing in around this time. Those numbers are US centric. Your followup point that fewer smokers exist and therefore more proportion would be expected to trend upwards no doubt and it did. >"To a great extent, this is a function of straightforward math... Fewer people smoke today than in previous decades... because there are fewer smokers in the population, out of every 100 lung cancer patients, fewer will therefore be smokers. And that means more of them will be never-smokers." This is totally valid and so is diesel exhaust as others mentioned. VW diesel scandal was also in this period as a painful confounder but at the same time I actually discount VW being of significant factor because how many people really drove TDIs? >A 2017 study of 12,103 lung cancer patients in three representative U.S. hospitals found that never-smokers were 8% of the total from 1990 to 1995 but 14.9% from 2011 to 2013. The authors ruled out statistical anomalies and concluded that “the actual incidence of lung cancer in never smokers is increasing.” Another study that same year, of 2,170 patients in the U.K., found an even larger increase: The proportion of lung cancer patients who were never-smokers rose from 13% in 2008 to 28% in 2014. From your link ^^^ About the time of that 2017 study from your article was when oncologists started freaking out by the shift in the new data. It has been getting worse each year. It's way too fast to be just a shift in statistics. It's also not a mystery neither. Obviously they spent the last ~5 years figuring it out.
- Gunax 4y ago
  'portion of cancer incodence in never-smokers' seems like a bad number to use. Clearly if we completely eliminated smoking, then the number would be 100%. It's clear to me that you and the doctors know that... But then why is this metric even mentioned? 20% versus 9% women/men... Are men just smoking more? It all feels like smoke and mirrors to me.
  - incomingpain 4y ago
    >It's clear to me that you and the doctors know that... But then why is this metric even mentioned? I think we all agree as well that tobacco smoking is bad and should stop. But not only are we never getting there, we also know this isn't the sole source of lung cancer or cancer in general. It's even possible that some new chemical or something else anytime in history might show up and start causing cancer far worse than tobacco. This is a proactive measurement that is necessary ongoing the rest of history or at least until we eliminate cancer in some manner. We also ultimately always want to be trending downwards. It means we know what we are doing and are improving the situation. If the situation is regressing, that's fine but as expected the appropriate people analyzed WHY. The answer they found... politically problematic. Same reason I'm flagged and unable to post on HN now.
    - manwe150 4y ago
      I have seen you say the answer is problematic, and obvious, but I am not sure I have actually seen you point to what that answer is. I think you partly might be flagged for forgetting to write your main point and just jumping right into defending it.

incomingpain 4y ago

I appreciate the response. I probably won't be able to reply, but I guess i'll be wasting my time typing a response. Sorry if I dont put as much effort in; sourcing etc. >A couple of data points. It looks like 20% is only in women - in men, it's 9%. Second, the ratio of smokers:nonsmokers can be at least partially explained by the number of smokers decreasing. Third, it's not clear whether this is a real change, or if it's a data anomaly. This is something I didn't fully understand and had to research. Why does it disproportionately affecting black people and women more? Nobody really explains it but the answer is known. Yes absolutely, as # of smokers decreases we would expect it to increase. Totally, but the correlation between decrease in tobacco usage and increase in lung cancer inverted. I think that's why when we saw an increase in 2006ish nobody responded like they have been responding in the last 5 or so years. >Personally, I would be very careful about the confidence with which I express this perspective. It might be true, but it's not at all clear that it is, so it's not clear there is a mystery. I hear your threat and yes I'm flagged and such. >“It is well-documented that approximately 20% of lung cancer cases that occur in women in the U.S. and 9% of cases in men, are diagnosed in never-smokers,” I believe what you are referencing is: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2764546/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2764546/ This study was done in 2006 publishing in 2007. Yes absolutely. 20% vs 9% was certainly a thing in around this time. Those numbers are US centric. Your followup point that fewer smokers exist and therefore more proportion would be expected to trend upwards no doubt and it did. >"To a great extent, this is a function of straightforward math... Fewer people smoke today than in previous decades... because there are fewer smokers in the population, out of every 100 lung cancer patients, fewer will therefore be smokers. And that means more of them will be never-smokers." This is totally valid and so is diesel exhaust as others mentioned. VW diesel scandal was also in this period as a painful confounder but at the same time I actually discount VW being of significant factor because how many people really drove TDIs? >A 2017 study of 12,103 lung cancer patients in three representative U.S. hospitals found that never-smokers were 8% of the total from 1990 to 1995 but 14.9% from 2011 to 2013. The authors ruled out statistical anomalies and concluded that “the actual incidence of lung cancer in never smokers is increasing.” Another study that same year, of 2,170 patients in the U.K., found an even larger increase: The proportion of lung cancer patients who were never-smokers rose from 13% in 2008 to 28% in 2014. From your link ^^^ About the time of that 2017 study from your article was when oncologists started freaking out by the shift in the new data. It has been getting worse each year. It's way too fast to be just a shift in statistics. It's also not a mystery neither. Obviously they spent the last ~5 years figuring it out.

Gunax 4y ago
'portion of cancer incodence in never-smokers' seems like a bad number to use. Clearly if we completely eliminated smoking, then the number would be 100%. It's clear to me that you and the doctors know that... But then why is this metric even mentioned? 20% versus 9% women/men... Are men just smoking more? It all feels like smoke and mirrors to me.
- incomingpain 4y ago
  >It's clear to me that you and the doctors know that... But then why is this metric even mentioned? I think we all agree as well that tobacco smoking is bad and should stop. But not only are we never getting there, we also know this isn't the sole source of lung cancer or cancer in general. It's even possible that some new chemical or something else anytime in history might show up and start causing cancer far worse than tobacco. This is a proactive measurement that is necessary ongoing the rest of history or at least until we eliminate cancer in some manner. We also ultimately always want to be trending downwards. It means we know what we are doing and are improving the situation. If the situation is regressing, that's fine but as expected the appropriate people analyzed WHY. The answer they found... politically problematic. Same reason I'm flagged and unable to post on HN now.
  - manwe150 4y ago
    I have seen you say the answer is problematic, and obvious, but I am not sure I have actually seen you point to what that answer is. I think you partly might be flagged for forgetting to write your main point and just jumping right into defending it.