Urban-rural differences in smoking prevalence in Greece.

Dear Editor, After perusing the report by Vardavas and Kafatos, on smoking policy and prevalence in Greece, we thought that would be of interest to demonstrate the findings of our study concerning the urban-rural differences in smoking prevalence. The study was conducted during the recent election day (15 October 2006) in the municipality of Salamis and Nemea. The first represent an urban area and the second a rural one. Data were collected through face to face interviews (response rate: 88% and 83%, respectively) conducted by trained health professionals. A current smoker was considered a person who was smoking at least one cigarette daily. Additional details of the samplingmethod are described elsewhere. The sample of urban and rural population consisted of 3478 adults (1731 males; 1747 females) and 976 (515 males; 461 females), respectively. Between two populations, as expected, there were significant differences in age and educational level. We used Pearson Chi-square test and binomial logistic regression analyses, applied in the total sample and within subgroups of demographic variables, to determine the urban-rural differences in the percentage of current smokers. Overall 39.1% (1742/4454) of the participants (48% of males, 30% of females), from both areas, were current smokers and 14.2% former smokers. As shown in table 1, the multivariate analyses fail to show significant urban-rural differences in the prevalence of smoking in total population, as in each sex. On the contrary, important differences were unmasked across age and education subgroups. In particular, the rates of smoking in the 20–50 age group were lower in urban than rural population, whereas an opposite pattern was found in age group of 51–97 years. In regard with educational level, we found that among lower educated people those living in the urban area smoke more than those living in the rural area. While among higher educated people the urban residents smoke less than rural counterparts. Previous studies from developed countries show that inhabitants (of both sexes) of urban areas aremore likely to be current smokers than inhabitants of rural areas, whereas in les developed countries such as Russia the studies show marked urbanrural differences only for women. As the present study show, in Greece, the pattern of urban-rural differences seems to be more complex, indicating that this is likely to be related with the stage of the smoking epidemic. In conclusion, the present study lending further power to the suggestion that for the effective intervention programs against smoking it is necessary to recognize and take into account the urban-rural differences in smoking prevalence of the particular country. And finally, we are adding to the suggestions of Vardavas and Kafatos that a national anti-smoking policy movement should firstly mobilize the physicians, who represent the tip of the spear in fight against the smoking.