Neurology

Secondhand smoke linked to dementia

"Passive smoking can dramatically increase the risk of developing severe dementia," is the Daily Mail’s headline. The newspaper reports that a study has been the first to show a significant link between exposure to secondhand smoke and dementia.

This news is based on a large Chinese study that looked at the "mental state" of over-60s and their lifetime exposure to secondhand smoke.

Researchers used a computer program to see whether the participants had problems with memory, thinking and changes in personality that were consistent with "moderate or severe dementia syndromes". They then looked to see whether there was a link between secondhand smoke exposure and dementia symptoms and found that people reporting any secondhand smoke exposure had a 29% increased risk of a "severe dementia syndrome" compared with those with no exposure to smoke.

The harms of passive smoking (such as increased risk of heart disease) are well-established and there could be a case to add dementia to the list in the future. However, this study cannot prove that this is the case, due to a number of limitations, including the uncertain reliability of both the dementia diagnoses and the recall of exposure to secondhand smoke. Also, while the researchers accounted for various potential confounders such as age and socio-economic factors, there is the possibility that other unmeasured factors are influencing the association. The results in this Chinese population may also not be applicable to people in the UK.

A definite link between dementia as a whole (not a specific type such as Alzheimer's or vascular dementia) and passive smoking cannot be reliably concluded from this study. Further research is required on this topic.

Where did the story come from?

The study was carried out by researchers from the University of Liverpool, the University of Oxford, the University of Wolverhampton, King’s College London and the University of Texas in the USA. The research was funded by the Alzheimer’s Research Trust and BUPA Foundation, and an individual researcher received support from the University of Wolverhampton Strategic Research Development Fund.

The study was published in the open access peer-reviewed medical journal Occupational and Environmental Medicine.

Passive smoking is known to be bad for health in general, so the Mail can perhaps be forgiven for its sensationalist headlines. However, due to the numerous limitations of this study it does not prove that secondhand smoking directly, and dramatically, increases risk of severe dementia.

What kind of research was this?

This was a cohort study that assessed almost 6,000 older adults in China, examining their mental state to see if they had dementia and questioning their exposure to secondhand smoke during their lifetime.

Secondhand or passive smoking has been associated with increased risk of cancer and cardiovascular diseases.

The researchers say that previous studies have also shown that smoking increases the risk of Alzheimer’s and other types of dementia, leading to speculation that passive smoking may carry a similar risk. However, no studies have yet tried to see how the level of exposure to secondhand smoke (what researchers would call the dose) may be linked to risk of dementia, and this is what they aimed to investigate.

There are several different types of dementia. Alzheimer’s disease is the most common, followed by vascular dementia. While vascular dementia is associated with cardiovascular disease (and therefore links to smoke exposure are quite plausible), the causes of Alzheimer’s – aside from increased age and possible genetic links – remain largely unknown. 

What did the research involve?

During 2007-9 the researchers chose one urban and one rural community from four different provinces in China and randomly selected roughly 500 individuals from each community.

Eligible adults had to be at least 60 years old and have lived in the area for five years.

Participants were interviewed at home using a general health and risk factors questionnaire and the Geriatric Mental State Examination (GMS). A computer program was used to identify the mental conditions affecting participants.

GMS symptoms were grouped into 150 “symptom components”, which helped the researchers group people into different diagnostic groups:

  • depression
  • schizophrenia
  • obsession
  • phobia
  • hypochondria
  • general anxiety
  • ‘organic’ disorders (organic generally means that there is a definite disease process going on in the brain which is responsible for symptoms – the most common type of organic brain disorder is dementia)

The program gives a number representing the level of confidence that a certain person has a particular diagnosis (0-5). They defined a person with 1-2 confidence level organic disorder as having a “moderate dementia syndrome” and levels 3–5 as “severe dementia syndromes”.

They used a questionnaire to assess the participants’ smoking history and secondhand smoke exposure. Current smokers were those who gave a positive answer to the question “Do you smoke cigarettes now?” and gave extra information about the duration of their habit and how many cigarettes they smoked each day. They also defined past smokers and those who had never smoked. All participants were also required to provide the answers “yes” or “no” relating to secondhand smoke exposure.

Respondents were asked about whether they experienced no, some or lots of secondhand exposure at home, in the workplace and in other places. All participants were asked how many years they were exposed to each of three sources of secondhand smoke.

When looking at links between secondhand smoke and dementia syndromes researchers adjusted their analysis for various potential confounders, including:

  • age
  • sex
  • smoking status
  • urban or rural location
  • educational level
  • occupational class
  • annual income
  • marital status
  • religion
  • current alcohol drinking
  • visiting children or relatives – which would suggest a good level of social support
  • hypertension
  • stroke
  • depressive syndromes

What were the basic results?

The study included 5,921 participants, 36% of whom had been exposed to secondhand smoke at some time (31% of never smokers and 46% of ex- or current smokers). Those previously exposed to secondhand smoke tended to be younger, have previously smoked themselves, live in rural areas, be of lower education or occupational class and drink alcohol. Moderate dementia syndromes (level 1–2) affected 14.1% of those with no secondhand smoke exposure and 15.7% with exposure. Severe dementia syndromes (level 3–5) affected 8.9% of those with no secondhand smoke exposure and 13.6% of those with exposure.

The researchers carried out numerous calculations according to location of exposure and duration of exposure. No significant associations were found for moderate dementia syndromes. For severe dementia syndromes they found that any secondhand smoke exposure increased the risk of severe dementia syndrome by 29% (relative risk 1.29, 95% confidence interval 1.05 to 1.59).

They found a general trend for a dose-dependent relationship, where increasing levels of exposure, ranging from 0-25 years to 100 or more years, increased the risk of severe dementia syndromes, though not all of these individual associations were significant.

How did the researchers interpret the results?

The researchers conclude that secondhand smoke exposure “should be considered an important risk factor for severe dementia syndromes”, and that avoidance of secondhand smoke “may reduce the rates of severe dementia syndromes worldwide”.

Conclusion

Secondhand or passive smoking is known to be detrimental to health and has been associated with increased risk of cardiovascular diseases and cancer. However, it is not yet certain whether it is linked to dementia.

Though this large study finds a significant link between secondhand smoke exposure and severe dementia syndromes there are several important limitations to be aware of.

Problems with measuring dementia diagnoses

The method for diagnosis of dementia used by this study was unusual. Though the researchers assessed each individual using a mental state examination, they went on to diagnose them using a computer program which assigned symptoms into different diagnostic groups or “syndromes”. They then attached a level of confidence that this syndrome was the correct diagnosis, from 0-5. Those who fell into the syndrome of “organic disorders” where classed as having dementia. An organic brain disorder usually means a condition where there is a definite disease process going on in the brain which is responsible for symptoms.

While types of dementia are the most common and largest group of organic brain disorders, this group would normally be expected to include other conditions caused by brain disease, damage or dysfunction. Therefore, it is difficult to say how accurate it would be to categorise all people in this syndrome group as having dementia. Also, how well the 1–2 level confidence can directly apply to moderate dementia, and 3–5 applies to severe dementia, is difficult to say. Mental state examination and cognitive tests often pick up more minor degrees of memory loss and it is not reasonable to assume that all memory changes are dementia or will inevitably lead to it. The most effective way to diagnose organic disorders would be to carry out a thorough clinical assessment, including the use of brain scans, but this option may not have been available to the researchers. Overall, it is difficult to say whether the diagnoses of dementia made were entirely accurate.

Reliability of smoke exposure reporting

Similarly there may be questionable reliability over the responses on exposure to secondhand smoke, particularly when participants were being asked to specify where they had been exposed and for how long.

Unaccounted confounding factors

Though the researchers have adjusted their analyses for several potential confounders, it is still difficult to conclude for certain that smoke exposure is directly responsible for any link seen, and that other unmeasured factors are not involved.

Population demographics

The study has only assessed a Chinese population. It may be difficult to apply these results to people of different cultures and ethnicities who have different exposure to secondhand smoke, and who are affected by other factors that could influence the risk of dementia.

The harms of passive smoking are well-established. However, whether there is a definite link between dementia as a whole (rather than a specific type of dementia) and passive smoking cannot be reliably concluded from this study.

As the researchers point out, currently 93% of the world’s population live in countries that do not have public smoking bans. Further research highlighting the potential dangers of passive smoking would be useful.


NHS Attribution