The New Straits Times, April 3, 1998
Q: I CAME across a news article that claimed that "The World's leading health organisation, World Health Organisation (WHO), has withheld from publication a study which shows that not only might there be no link between passive smoking and cancer but that it could even have a protective effect."
How true is this statement?
A: THE above allegation was based on a study conducted by WHO which looked at the effect of environmental tobacco smoke on the risk of getting lung cancer in European populations.
This study was carried out over the last seven years by 12 research centres in seven European countries under the leadership of WHO's cancer research branch - the International Agency for Research on Cancer (IARC).
Based on this study, the Press statement as highlighted in the above question has claimed that WHO has withheld from publication its own report that was aimed at but supposedly failed to prove that there is an association between passive smoking or environmental tobacco smoke (ETS) and a number of diseases, lung cancer in particular.
Both statements were admitted as untrue judging from the Press release produced by WHO (a copy of the press release can be retrieved from WHO website.
The WHO-IARC study in question has been submitted to a scientific journal for evaluation and at present is undergoing peer review before being accepted for publication. The study showed a relative risk of 1.16 for a non-smoker contracting lung cancer as a result of living with a smoking spouse or 1.17 relative risk from working in a smoking workplace.
Taking into consideration these values, this study found that non-smokers are 16 per cent more likely to get lung cancer if their spouse smokes than if they live with a non-smoking spouse.
In practice this is a small risk as compared to an active smoker who has a risk of 20 times higher than a non-smoker but when applied to the millions of people under this situation, there would be an additional 16-17 per cent extra risk which would amount to an extra several hundred deaths per year due to environmental tobacco smoke.
A point to note in this case is that although the diference is not statistically significant, it could still be real. To illustrate this further, these data showed that there could have been a lesser incidence of lung cancer by as much as 16 per cent in favour of non-exposure to ETS, which would certainly by important if it turned out to be the case.
The second issue that was raised to show no association between passive smoking and lung cancer was in the interpretation of the statistical test applied to these results.
The opponents to the WHO-IARC report claim that there was no statistically significant correlation for the above statement was taken from the application of the significance test to interpret any comparative result.
So, the underlying test being used is a null hypothesis for which there is no diffrence in population parameters among the groups being compared. In other words, the null hypothesis is consistent with the notion that the observe difference is simply the result of random variation in the data.
The principle involved in such testing usually involves a computing of the test statistics using a standard equation and comparing it with a critical value obtained from a set of statistical tables. When the test statistics exceed the critical value, the null hypothesis is rejected and the difference is declared statistically significant.
In any decision to accept or reject the null hypothesis, a significance level of testing is normally accepted where the risk of rejecting the null hypothesis is stated even though the hypothesis is correct. For example, if the significance level is at five per cent, then there is a five per cent chance of rejecting the null hypothesis when it is true.
Although there is uncertainty at five per cent level, the use of a lower test of statistical significance such as 20 per cent level would certainly provide a statistically significant link between passive smoking and lung cancer in the WHO-IARC study.
Thus looking at the importance of this difference under a different context, it can be said that a policy decision by any government of no action towards passive smoking would be fiercely contested if it was 80 per cent sure that it causes lung cancer.
Since the estimate of risk in the WHO-IARC study is based on a sample of 650 lung cancer cases, the above-metioned risk to the whole population could be not statistically different as the sample may not be exactly representative.
The difference could be significant only if a bigger sample of lung cancer patients was used or the derived figures were compared with another larger study involving a bigger number of samples.
Based on this argument, a comparative study with another larger study of 4,626 cases of lung cancer (British Medical Journal Vol 315 page 980-988, Oct 18 1997) showed that the derived ranges from WHO-IARC study overlap and therefore the results are consistent.
The relative risk from the larger study showed that the significant exposure to environmental tobacco smoke was 1.44 or 44 per cent more likely to develop lung cancer as compared to non-exposed population.
As such, these two independent research indicate that there is a link between passive smoking and lung cancers. In retrospect, the opponents of WHO-IARC study have inverted this uncertainty to a rather limited view of accepting the study as evidence of no effect or the risk is insignificant.
Besides the epidemiology data, there are also sources of evidence that fully support the above argument. To conclude, an utmost important point to take note from the WHO-IARC study as well as other studies conducted elsewhere is PASSIVE SMOKING CAUSES LUNG CANCER IN NON-SMOKERS. - National Poison Centre.
