front |1 |2 |3 |4 |5 |6 |7 |8 |9 |10 |11 |12 |13 |14 |15 |16 |17 |18 |19 |20 |review |
We used multiple
logistic regression to examine the determinants of current smoking with never and ex-
smokers combined as the reference category. The determinants were considered in a number
of domains which have been found to have important influences on health and behaviour in
the former Soviet Union: Sociodemographic factors: age, urban/rural status, region of residence, education, occupation and nationality. Material factors: income, self rated material situation and unemployment. Psychosocial factors: Social position, change in social position over the last 5 years and involvement in social networks measured using marital status, church attendance and communist party membership. Traditional measures of socio-economic status used in the west cannot be assumed to apply in the post-Soviet context. For example, education is not as clear a marker of material status as it is in the west. For this reason, measures of social class and education were considered as demographic characteristics and under material status we considered only variables that more directly measure material position. Age adjusted gender specific odds ratios were calculated for each variable. Then, to allow for potential confounding, a multivariable analysis was performed. The results of this multivariable analysis are presented in the table. |
front |1 |2 |3 |4 |5 |6 |7 |8 |9 |10 |11 |12 |13 |14 |15 |16 |17 |18 |19 |20 |review |