Since 1992, the standard socio-economic variable used to reflect living circumstances has classified all members of a household according to the present or last job of the head of the household: from 2000 the classification is based on the occupation of the household reference person. Prior to 1992, the method for assigning socio-economic group to GHS respondents involved classifying married or cohabiting women according to their partner’s present (or last) job. This earlier definition has been retained in Table 8.8 to permit long term analysis of change over time. A third socio-economic variable is based on each person’s own current or last occupation.

The GHS has consistently shown, with whichever socio-economic variable is used, that cigarette smoking is considerably more prevalent among those in manual groups than among those in non-manual groups.

• In the 1970s and 1980s, the prevalence of cigarette smoking fell more sharply among those in non-manual than in manual groups, so that differences between the groups became proportionately greater.

In the 1990s there was little further change in the relative proportions smoking cigarettes until 1998 to 2000, when there was a fall of two percentage points in the prevalence of smoking among both men and women in manual socio-economic groups.

• In 2000, 34% of men living in households in the manual group smoked cigarettes compared with 23% of those in non-manual households. The corresponding proportions for women were 29% and 22%.
• Men living in households in the unskilled manual group were more than twice as likely as those who lived in professional households to smoke: 39% did so, compared with 17%.

In England, the overall prevalence of cigarette smoking among those in manual socio-economic groups fell from 33% in 1998 to 31% in 2000. This is a statistically significant fall and in line with the NHS Cancer Plan target of reducing smoking among those in manual groups. However, caution is advisable when making this year on year comparison because the data may have been affected by the change from head of household to household reference person as the basis for assessing socio-economic group.