Author(s):
Leite, Andreia 
; Machado, Ausenda ; Dias, Carlos Matias
Date: 2013
Persistent ID: http://hdl.handle.net/10400.18/1709
Origin: Repositório Científico do Instituto Nacional de Saúde
Subject(s): Determinantes da Saúde e da Doença; Sociodemographic Factors; Tobacco Consumption and Cessation; National Health Interview Survey; Multinomial Log-linear Model
Description
Background: Smoking is a significant risk factor for several diseases.
Social inequalities have been described for tobacco consumption and
though to a lesser extend for tobacco cessation.
Objectives: Describe sociodemographic factors associated with
tobacco consumption and cessation.
Methods: Data from the 2005/2006 Portuguese National Health
Interview Survey was analysed. A multinomial log-linear model was
fitted considering 3 groups: present smokers, former smokers and
never smokers. Relative Risk Ratios (RRR) were obtained. Men and
women were analysed separately. RRR adjusted for age and comparatively
with reference categories (Centre for regions, married for
marital status, student for professional situation and higher education
for education) are presented. RRR of present smokers compare to
never smokers and RRR of former smokers to present smokers.
Results: Men from groups that have higher RRR of present smoking
also have lower RRR of stopping. The lowest risk of cessation and
higher of consumption was observed in Azores [RRR (95 % Confidence
Interval (CI 95)) 0.53 (0.46, 0.61) and 1.89 (1.64, 2.18),
respectively]. Divorcees, unemployed, and men with lower secondary
education all had lower risk of cessation and higher of consumption,
[respectively RRR (IC 95): 0.50 (0.41, 0.61) and 2.01 (1.64, 2.46);
RRR (IC 95): 0.16 (0.13, 0.20) and 6.29 (5.00, 7.91); RRR (IC 95):
0.65 (0.56, 0.77) and 1.53 (1.30, 1.79)].
Similar results were observed in women (higher RRR of smoking
and lower RRR of stop). Exceptions were found in the non-unemployed
women and those with less that upper secondary. Taking
region into consideration women in Madeira had a lower RRR of stop
smoking and in Lisbon they had the higher risk of smoking [RRR (IC
95) 0.40 (0.27, 0.58) and 2.11 (1.74, 2.57), respectively]. The same
was observed for divorced women [RRR (IC 95) 0.54 (0.41, 0.69) and 3.21 (2.69, 3.84), respectively]. Unemployed had the lower risk of
quitting smoking and the highest of smoking [RRR (IC 95) 0.83 (0.51,
1.38) and 4.23 (3.20, 5.58), respectively], while the other groups had
highest RRR of stop smoking and smoking, comparatively with reference
group. Less educated women (lower secondary or less) had
lower risk of cessation but also lower risk of smoking.
Conclusions: Unfavourable sociodemographic characteristics are
associated with higher risks of being a current smoker and lower risks
of being an ex-smoker. Different/further cessation measures are
needed in these less prone to quit group of individuals, namely men
from Azores, divorced, unemployed and with lower secondary education
and for divorced and unemployed women.
