Application of Negative Binomial Regression for Assessing Public Awareness of the Health Effects of Nicotine and Cigarettes

Abstract

Both the public and private sectors have acted responsibly to help decrease smoking-related deaths by putting health warnings on all cigarette packages. This study investigated the social or demographic factors associated with public awareness of health warnings on the harmful effects of environmental tobacco smoke based on baseline data collected by the South African Human Sciences Research Council (HSRC). Respondents in the survey were asked to recall the number of anti-smoking messages which appeared as warning messages on cigarette advertisements. The number of anti-smoking messages recalled ranged from 0 to 9 with a mean of 3.09 (variance of 5.99) and a median of 3.00. Because the variance was nearly two times greater than the mean, the negative binomial regression model provided an improved fit to the data and accounted better for overdispersion than the Poisson regression model, which assumed that the mean and variance are the same. The level of education and race were found to be the most significant factors. Moreover, the lower socio-economic class non-smokers’ anti-smoking messages recalling rate was 2.5 times that of the lower socio-economic class smokers. Unlike men, women’s anti-smoking message response rate increased with income.