THE RELATIONSHIP BETWEEN CIGARETTE-SMOKING AND EDUCATION REVISITED - IMPLICATIONS FOR CATEGORIZING PERSONS EDUCATIONAL STATUS

Objective. This study sought to reassess the relationship between ciga rette smoking and education. Methods. Data from the 1983 to 1991 Natio nal Health Interview Survey for participants aged 25 years and older w ere used to plot the prevalence of current smoking, ever smoking, heav y smoking, and smoking cessation, as well as the adjusted log odds rat ios, by years of education. Results. The ''less than high school gradu ate'' category consisted of two groups with distinct smoking patterns: persons with 0 to 8 years and persons with 9 to II pars of education. The latter were the most likely to be current, ever, and heavy smoker s and the least likely to have quit smoking, whereas the former were s imilar to persons having 12 years of education. After II years of educ ation. the likelihood of smoking decreased and that of smoking cessati on increased with each successive year of education. These results per sisted after the statistical adjustment for age, sex, ethnicity, pover ty status, employment status, marital status, geographic region, and y ear of sun?ey. Conclusions. The relationship between smoking and educa tion is not monotonic. Thus, when evaluating smoking in relation to ed ucation, researchers should categorize years of education as follows: 0 to 8, 9 to 11, 12, 13 to 15, and 16 or more years.