Skip to main content

Smoking, class and sex

Wednesday 18 February 2009

“Smoking kills regardless of class,” BBC online has reported. The website says that a new study has shown that “being female or rich offers no defence against the ill heath caused by smoking”. The study followed 15,000 people from Scotland over a 28-year period, comparing gender, class, smoking habits and survival rates.

The study found that “smokers of all social classes had a much higher risk of premature death than even the poorest non-smokers”. The study also suggested that the survival rates of people that gave up smoking in the long-term were closer to those of people who had never smoked than to current smokers. The authors say their work provides further evidence that “cigarettes indiscriminately damage and kill their users, regardless of social position”.

This large study has provided more evidence about the dangers of smoking, and illustrates that these dangers significantly affect people of all social classes. These results show that social status cannot protect against smoking-related health problems, and should provide further incentive for all smokers to give up.

Where did the story come from?    

Dr Laurence Gruer and colleagues from NHS Health Scotland and the University of Scotland carried out this research. The analyses in this study were funded by NHS Health Scotland, and the original study was funded by King Edward Memorial Fund and the Scottish Home and Health Department. The study was published in the peer-reviewed British Medical Journal.

What kind of scientific study was this?    

This was a prospective cohort study that looked at the effect of smoking on survival among people of different social status.

Between 1972 and 1976 researchers asked all adults aged 45 to 64 years in Renfrew and Paisley, west-central Scotland, to participate in the study. In total 15,402 people, about 80% of those invited, were recruited into the study.

Participants filled out a questionnaire about themselves, including smoking habits, area of residence and occupation. Retired people gave their previous occupation, and housewives gave their husband’s occupation.

People’s social class was determined by their occupation, based on a standard system. Class groups were organised into four levels: highest (classes I and II), class III non-manual, class III manual, and lowest (classes IV and V).

The researchers also estimated a second, separate measure of social position by using a standard classification system to determine how deprived each participant’s local area was. This classification system uses seven categories, with a higher number indicating greater deprivation. Within this study the researchers further grouped these seven categories into four groups: most affluent (groups 1 to 3), group 4, group 5 and least affluent (groups 6 and 7).

Based on their questionnaire response, people were classed as current smokers (smoking within the past year), former smokers (stopped smoking at least a year ago) or never smoked. Participants also had a physical examination that included measuring their height and weight, lung capacity, blood pressure and cholesterol level.

The researchers followed up participants for 28 years, collecting data on deaths from the General Register Office for Scotland. People stopped contributing data once they left the UK. To analyse the data, participants were split into 24 non-overlapping groups based on gender, smoking status, social class, deprivation category and year of entry to the study.

Statistical analysis was used to look at relative death rates among smokers in the different social classes and deprivation categories. Analyses took into account factors that could affect results, such as age, blood pressure, body mass index, cholesterol levels and lung capacity. Because death rates increase as people get older, the analyses of the 28 years of follow-up were split into two consecutive 14-year periods.

What were the results of the study?

Of the 15,402 people recruited, 14,955 provided complete data (97%). Within each of the social class groups, 43% to 52% of women smoked at the start of the study. The highest rate of smoking was among the lowest class group, and the lowest rate was among the highest class group.

Between 47% and 64% of men in each social class smoked at the start of the study. Men in the lowest social class group had the highest rate of smoking, and men in the highest social class group had the lowest rate. Over the 28 years of follow-up, 55% of the women and 70% of the men died. 

Among women who had never smoked, 65% in the highest social class and 56% in the lowest social class were alive after 28 years (once adjusted for differences in participants’ ages). The rates of survival at 28 years were lower among women who were current smokers at the start of the study (41% in the highest social class, 35% in the lowest social class).

The same pattern of survival was seen among men at 28 years: survival rates were 53% of those who had never smoked in the highest social class, 36% among those who had never smoked in the lowest social class; 24% among current smokers in the highest social class and 18% among current smokers in the lowest social class.

In both men and women, survival rates among those who had never smoked from the lowest social classes were much better than among smokers in the highest social classes. Similar findings were reported through analysis using deprivation class as a measure of social standing. Women who smoked had lower survival rates than men who had never smoked in all except the lowest social class groups.

Women from the highest social classes who had never smoked had the lowest proportion of deaths. The likelihood of death in each group was reported in relation to this group. Compared with the group:

  • Women who currently smoked at the start of the study were between 1.7 and 2.5 times more likely to die during follow-up, depending on their social class.
  • Women who were former smokers were between 1.4 and 2.4 times more likely to die during follow-up.
  • Men who had never smoked were between 1.7 and 2.2 times more likely to die.
  • Current male smokers at the start of the study were between 3.5 and 4.2 times more likely to die.
  • Former male smokers at the start of the study were between 2.1 and 2.7 times more likely to die.

The researchers reported that they had similar findings through analysis of survival based on deprivation class of the area in which the participants lived.

What interpretations did the researchers draw from these results?     

The researchers conclude that people who have never smoked have much better survival rates than smokers in all social classes. Smoking was a greater source of differences in death rates than social class, and eradicated women’s survival advantage (which means women generally have longer lives than men). The authors say that “this suggests the scope for reducing health inequalities related to social position in this and similar populations is limited unless many smokers in lower social positions stop smoking”.

What does the NHS Knowledge Service make of this study?     

This large study provides information about the long-term survival rates among smokers and non-smokers in different social classes and deprivation categories. The study’s strengths included the high rate of participation, and high proportion of participants providing complete data. In addition, analysing data using two different measures of social position also increases confidence in the study’s findings.

There were also some limitations to the research, some of which the authors discuss:

  • This study only collected information about smoking at the start of the study, but some people may have changed their smoking behaviours over the period of the study, and this could have affected the results. The authors report that, based on trends in smoking in Scotland, it is likely that a considerable proportion of participants stopped smoking after entering the study.
  • The authors report that occupation is a relatively weak measure of a person’s social class. This is particularly the case for women who do not work, who were categorised into a social class based on their husband’s occupation.
  • In their analysis, the researchers adjusted for some possible confounding factors, which increases the reliability of the results. However, these adjustments may not have removed confounding completely. Unknown or unmeasured factors might affect results.
  • The authors report that there were relatively few people from the higher social classes in their study, which meant that they had to group some of the higher classes together. This means that they could not separate out the impact of smoking within these higher social classes.

This study highlights the fact that smoking remains a significant risk factor for people of all backgrounds, even more affluent individuals. The results should provide further incentive to those who smoke to give up, regardless of their social class.

Analysis by Bazian
Edited by NHS Website

Links to the headlines

Smoking kills regardless of class.

BBC online, 18 February 2009

Links to the science

Gruer L, Hart CL, Gordon DS, Watt GCM.

Effect of tobacco smoking on survival of men and women by social position: a 28 year cohort study.

BMJ 2009;338;b480