Dutch survey: One in four heavy smokers never reaches pension age

[Palinurus]

Source: https://www.dutchnews.nl/news/archives/2017/09/one-in-four-heavy-smokers-never-reach-pension-age-survey-shows/

One in four heavy smokers never reaches pension age, Dutch survey shows

September 15, 2017

One out of every four smokers dies before their 65th birthday, according to a new survey published by the Trimbos institute of mental health and addiction, and the national statistics office CBS.

The study of the relationship between smoking and death involved 40,000 respondents aged between 20 and 80 from 2001 and 2006, and was based on their dates of death.

The research revealed that smokers died relatively young: 23% of heavy smokers (more than 20 cigarettes a day) never reached the age of 65. A full 11% of light smokers died before they reached 65 as did 7% of non-smokers.

The life expectancy of heavy smokers was on average 13 years shorter than that of non-smokers. Moderate smokers (fewer than 20 cigarettes a day) lost an estimated nine years, while light smokers who did not smoke on a daily basis had a five year lower life expectancy, the study revealed.

Smokers die from lung cancer relatively often, but heart, vascular diseases and respiratory problems are increasingly frequent causes of death.

Stopping smoking pays off at all ages, the study revealed. Former smokers who stop before their 35th birthday have a similar life expectancy to those who never smoked. People who stop at 50, halve their risk of an early death. Nevertheless four out of every 10 people who die before the age of 80, are killed by the effect of tobacco.

 

[Ant22]

The survey doesn't make any distinction between smokers who smoke commercial cigarettes with harmful chemicals and additives in them and those who smoke organic tobacco. Would it be fair to say that if the survey was divided into commercial cigarette smokers and organic tobacco users the results would be different?

 

[Palinurus]

Most smokers over here are not aware of that distinction either, so I don't suppose results would differ substantially when such a division was implemented -- if such indeed were practically possible since there are no hard facts or statistics about this distinction available AFAIK.

 

[Oxajil]

I found the paper here (in Dutch): https://www.ntvg.nl/artikelen/doodsoorzaak-nr-1-bij-jonge-nederlanders-de-sigaret/volledig

Their method was:

1. Surveys (2001-2006) about 'smoking behavior' among 40,000 20-79 year-old participants were linked to cause of death registries.
2. Of these participants, they checked whether they had died within 120 months (10 years) after filling out the survey.
3. If participants passed away, they looked at the precise date of death and the underlying cause of death.

As far as I can see, there is no correction made for any factors, such as lifestyle, for example.

Their conclusion that "4 in 10 premature deaths in the Netherlands is caused by smoking" is a stretch, since their results only say that heavy smokers may die at an earlier age (which can be due to many factors, not included in the study), and not that it is smoking that is causing premature death.

I could be wrong, but that's how it seems to me.

 

[Zar]

I found the paper here (in Dutch): https://www.ntvg.nl/artikelen/doodsoorzaak-nr-1-bij-jonge-nederlanders-de-sigaret/volledig

Their method was:

1. Surveys (2001-2006) about 'smoking behavior' among 40,000 20-79 year-old participants were linked to cause of death registries.
2. Of these participants, they checked whether they had died within 120 months (10 years) after filling out the survey.
3. If participants passed away, they looked at the precise date of death and the underlying cause of death.

As far as I can see, there is no correction made for any factors, such as lifestyle, for example.

Their conclusion that "4 in 10 premature deaths in the Netherlands is caused by smoking" is a stretch, since their results only say that heavy smokers may die at an earlier age (which can be due to many factors, not included in the study), and not that it is smoking that is causing premature death.

I could be wrong, but that's how it seems to me.

I'm not an expert but when it comes to smoking statistics there is something that I am curious about. In this upside down society smoking is considered detrimental for the health, so any person who cares about their health should not smoke. That being said we can conclude that smokers should care less about their health and be more susceptible to diseases, hence the higher death rates... So as you pointed out Oxajil taking into account their lifestyle is fundamental to getting objective data. I'd file this under the "shady statistics" department, IMO.

 

[Palinurus]

As far as I can see, there is no correction made for any factors, such as lifestyle, for example.

According to the article there seems to be a reason for that:

We vergeleken de verschillende categorieën rokers met de nooit-rokers wat betreft risico op voortijdig overlijden, uitgesplitst naar de belangrijkste doodsoorzaken. De hazardratio (HR) op voortijdig overlijden voor rokers werd bepaald met een naar geslacht en leeftijd (continu) gestratificeerd Cox-‘proportional hazards’-model.4 Onderzocht is of het verband tussen voortijdig overlijden en rookgedrag deels veroorzaakt wordt door andere factoren waarin rokers verschillen met nooit-rokers, bijvoorbeeld op het gebied van leefstijl. Correctie voor persoonskenmerken bij aanvang van het onderzoek – alcoholgebruik, lichamelijke activiteit, opleidingsniveau, BMI of een combinatie hiervan – had echter nauwelijks of geen significant effect op de HR. Voor dit artikel rekenden we dan ook verder met de ongecorrigeerde HR.

We compared the different categories of smokers with the never-smokers regarding risk of premature death, broken down to the main causes of deaths. The hazard ratio (HR) on premature death for smokers was determined with a for sex and age (continuous) stratified Cox-'proportional hazards'-model.4 Examined was whether the relationship between premature death and smoking behavior is partly caused by or is due to other factors in which smokers differ with never-smokers, for example in the field of lifestyle. Correction for personal characteristics at the start of the investigation – alcohol consumption, physical activity, level of training, BMI or a combination of these – had hardly or no significant effect on the HR. For this article, we subsequently continued with the uncorrected HR.

Their conclusion that "4 in 10 premature deaths in the Netherlands is caused by smoking" is a stretch, since their results only say that heavy smokers may die at an earlier age (which can be due to many factors, not included in the study), and not that it is smoking that is causing premature death.

Table 1 shows that mortality has already increased from the age of 30 as the participants were more heavily smoked; An extended version of table 1 is in Supplement 2. Thus, 6% (95%-BI: 5-8) died of the heavily smoking 40-49-year-olds within 10 years of the survey's uptake; Of the 40-49-year-olds who had never smoked this was 1% (95%-BI: 1-1). Of the heavy smokers in the age of 60-69 years 27% (95%-bi: 23-32) died within 10 years, compared with 8% (95%-bi: 7-10) of the never-smokers. Lung cancer was the most common cause of death among these relatively young deceased smokers.

Even beyond 69 years of age, there was still a difference between the smoker categories. Of the 70-79-year-old heavy smokers 61% (95%-bi: 52-69) died within 10 years and from the never-smokers 30% (95%-bi: 28-32) (see table 1). At this age, cancer and cardiovascular disease each contributed to a third of mortality, both smokers and never-smokers. Mind you, at a higher age, the primary cause of death is more difficult to determine because older people often suffer from multiple chronic conditions. 11

The HR's on death for smokers compared to never-smokers were for the 10-year sledding time groups of 30-69 years approximately equal: heavy smokers died 3.8 times (95%-BI: 3.2-4.5) more often within 10 years as their peers who had never smoked (table 2; An extended version of table 2 is in Supplement 3). In the moderate smokers, the HR is 2.6 (95%-bi: 2, 3-3.0) and the Light smokers 1.7 (95%-bi: 1, 3-3, 3)

Especially the difference in cancer death was great. The HR was 4.3 (95%-bi: 3.4-5.5) for heavy smokers and 3.1 (95%-bi: 2, 5-3.9) for the moderate smokers; The HR's on death from lung cancer were high (see table 2).

As someone smoked more cigarettes, the risk of early death of cancer or cardiovascular disease was particularly higher. A dose effect could not be demonstrated in respiratory diseases. However, smokers died relatively often from a respiratory disease, particularly to COPD.

In older age groups the difference in 10-years mortality between smokers and peers who had never smoked was lower. For example, the HR for 70-79-year-old heavy smokers was 2.5 (95%-BI: 1.9-3.2) (see table 2).

I'm not an expert but when it comes to smoking statistics there is something that I am curious about. In this upside down society smoking is considered detrimental for the health, so any person who cares about their health should not smoke. That being said we can conclude that smokers should care less about their health and be more susceptible to diseases, hence the higher death rates... So as you pointed out Oxajil taking into account their lifestyle is fundamental to getting objective data. I'd file this under the "shady statistics" department, IMO.

To date, the long-term prognosis of mortality risks and life expectancy in the Netherlands is based on international risk estimations for smokers on lung cancer. 13.14 The current analysis provides the first figures based on Dutch empirical data only and is also up-to-date.

The risk of an early death increased with rising smoke intensity and an increasing number of smoke years. Smokers who smoked their entire lives at least 20 cigarettes a day died almost 4 times more often at a relatively young age than those who had never smoked. In smokers who smoked permanently less than 20 cigarettes a day, the risk was 2.5 times higher; For smokers who did not smoke every day the risk was still doubled. Heavy smokers lost on average 13 years of life due to their smoking behavior, moderate smokers estimated 9 and light smokers 5.

Based on these hazard ratios and the smokers prevalence of 30-69-year-olds in our research group, it appeared that in recent years about 4 out of 10 premature deaths in the Netherlands were caused by cigarette smoking; 20 years ago this was estimated to be more than a quarter. 14 We now see the effects of the Dutch with the heaviest smoke past in history.

I think it would be proficient to read the entire paper in google translation (Bing translator cannot provide the whole lot due to https url which they don't cover).

 

[Oxajil]

Correction for personal characteristics at the start of the investigation – alcohol consumption, physical activity, level of training, BMI or a combination of these – had hardly or no significant effect on the HR. For this article, we subsequently continued with the uncorrected HR.

Then why continue with the uncorrected? It would have made a stronger impact if they had included all of that data in their analyses. It would've been good I think if they provided their raw data, and if they included all of that information, so that others could see as well.

I'm not sure what you were trying to say Palinurus with the quote to my second comment (which was that their article didn't prove that smoking is the direct cause of these premature deaths). One can make the observation that a group of heavy smokers dies at a younger age compared to a group of non-smokers, but you can't say that that result is conclusive evidence that smoking causes premature death. Considering the anti-smoking campaign that's been running pretty strongly for many years, I'm extra skeptical about reports such as this one.

 

[flashgordonv]

I must say I find these sorts of research quite predictable and tedious. This is another study based on epidemiological methodology - which is questionnaire based with data manipulated with statistics.

Now, the reality is that one can produce a questionnaire to get any answer you want. It all depends on
1. the questions you ask,
2. the questions you don't ask,
3 How you ask the questions, the slant you put on them,
4. the confounding factors you ignore.

The you need to consider that all of the so called "smoking related" illnesses and diseases, are diseases with multiple risk factors - cancer, COPD, heart disease as a common core of often quoted examples. Yet if you read the research on alcohol, on obesity, on sugar etc, you find all of these are supposed causes for the same diseases. There are so many potential causes of these common diseases - old age, environment, pollution, work, stress, genetics, diet, socio-economic situation - that is is almost impossible for an epidemiological study to do anything more than show a correlation between a particular activity and a disease. And as we all should know, correlation is not causation. (For example, I study basketball. I find that tall men play basket ball. That is a correlation. If I conclude that playing basket ball makes men tall - that is not a proven causation, it is just nonsense.)

I wrote an article on this subject some months ago. the link is below - worth a read in this context I think
https://www.sott.net/article/315356-The-epidemic-of-junk-science-in-tobacco-smoking-research

 

[JGeropoulas]

...
I'm not an expert but when it comes to smoking statistics there is something that I am curious about. In this upside down society smoking is considered detrimental for the health, so any person who cares about their health should not smoke. That being said we can conclude that smokers should care less about their health and be more susceptible to diseases, hence the higher death rates... So as you pointed out Oxajil taking into account their lifestyle is fundamental to getting objective data. I'd file this under the "shady statistics" department, IMO.

Reminds me of a study about vegetarians, who came out as "healthier" than non-vegetarians. But the flaw was the same kind you mentioned above: The study didn't control for other variables related to vegetarians, who often do other things consistent with their desire for optimal health, such as meditation or yoga.

 

[Phill4]

Heavy smoker who smoke to cope with anxiety or tha abuse other substances, have higher risks of developing other stuff. I mean, heavy smokers don't just smoke.

Part of the reason a person develops are "heavy" addiction to a substance may be for stress managing reasons, more than it is for fun or enjoyment. Stress can equally damage the organism, specially the heart. Maybe for them not smoking can lower their life expectancy even more?

What I mean is that most of these studies leave out other direct factors that impact people's health. Stress, diet, anxiety, and toxic environments etc.

 

[loreta]

Those studies are like those programs that they do for people to stop smoking. One reason is to make people that smoke feel guilty and bad and stressed. It is my idea, very personal.

I know it is a serious subject but I am unable not to put this video that laugh a little about people who stop smoking.

http://<iframe width="560" height="...iw" frameborder="0" allowfullscreen></iframe>

 

[Eulenspiegel]

If you smoke a pack of cigarettes everyday, then it will likely have a negative influence on your health. No matter if they're organic or not. So as with many things, you gotta factor in how balanced that habit is compared to your overall lifestyle.

 

[Azur]

Those studies are like those programs that they do for people to stop smoking. One reason is to make people that smoke feel guilty and bad and stressed. It is my idea, very personal.

I know it is a serious subject but I am unable not to put this video that laugh a little about people who stop smoking.

http://<iframe width="560" height="...iw" frameborder="0" allowfullscreen></iframe>

That was brilliant!

Thanks Loreta.

 

[Palinurus]

I wrote an article on this subject some months ago. the link is below - worth a read in this context I think
https://www.sott.net/article/315356-The-epidemic-of-junk-science-in-tobacco-smoking-research

It took me a while to reconsider, to re-evaluate and to adjust, and I have to say your article was of tremendous help in doing so. I'm sure I missed reading it first time round, otherwise I wouldn't have bothered to signal this Dutch survey as a possible article for SOTT. So, many thanks to you; first and foremost for writing it and then once more for taking the trouble to mention it here again in its proper context.

I have to confess that the Dutch paper seemed to me quite solid at first glance, which clearly indicates a serious gap in my knowledge base. However, when confronted with the long list of possible pitfalls, foibles and disingenuity from the junk science article it obviously couldn't stand up to thorough scrutiny anymore. Lesson learned. Take away slogan: correlation does not equate causation.

This junk science seems a very silly business once you're in-the-know of its serious flaws. Couldn't help myself thinking: parrots parroting parrots parroting parrots... And so on, ad infinitum.

Thanks to all others for their contributions.

 

[Oxajil]

I wrote an article on this subject some months ago. the link is below - worth a read in this context I think
https://www.sott.net/article/315356-The-epidemic-of-junk-science-in-tobacco-smoking-research

It took me a while to reconsider, to re-evaluate and to adjust, and I have to say your article was of tremendous help in doing so. I'm sure I missed reading it first time round, otherwise I wouldn't have bothered to signal this Dutch survey as a possible article for SOTT. So, many thanks to you; first and foremost for writing it and then once more for taking the trouble to mention it here again in its proper context.

I have to confess that the Dutch paper seemed to me quite solid at first glance, which clearly indicates a serious gap in my knowledge base. However, when confronted with the long list of possible pitfalls, foibles and disingenuity from the junk science article it obviously couldn't stand up to thorough scrutiny anymore. Lesson learned. Take away slogan: correlation does not equate causation.

This junk science seems a very silly business once you're in-the-know of its serious flaws. Couldn't help myself thinking: parrots parroting parrots parroting parrots... And so on, ad infinitum.

Thanks to all others for their contributions.

Yeah, it's a great article! And it's good to hear it helped you, Palinurus :) I've had many eye-opening experiences such as the one you describe! I'm glad you did post about it here, because it has helped you, and I think it has been or will be helpful to many others as well. Studies like these can be really sneaky. If you'd like to read more on the topic, the following ones may be helpful as well:

- A comprehensive review of the many health benefits of smoking Tobacco.
- Let's All Light Up!
- Nicotine - The Zombie Antidote