1. Alcohol control laws, inequalities and geographical clusters of hazardous alcohol use in Geneva, Switzerland
José Luis Sandoval1,2,3, Teresa Leão4, Rebecca Himsl1,3,5, Idris Guessous1,3,6
1) Unit of Population Epidemiology, Division of Primary Care Medicine, Department of Community Medicine, Primary Care and Emergency Medicine, 2) General Internal Medicine Department, Geneva University Hospitals, Geneva, Switzerland; 3) Group of Geographic Information Research and Analysis in Public Health (GIRAPH); 4) Escola Nacional de Saúde Pública, Universidade NOVA de Lisboa, Lisbon, Portugal; 5) Laboratory of Geographic Information Systems (LASIG), School of Architecture, Civil and Environmental Engineering (ENAC), Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland; 6) Department of Ambulatory Care and Community Medicine, University of Lausanne, Lausanne, Switzerland
Introduction
According to the WHO, 3.3 million deaths per year (5.9% of global mortality) are the result of harmful use of alcohol, representing a large economic, social and disease burden to societies.[1]
As other harmful behaviours, hazardous alcohol consumption is socially patterned. In high income countries men, but not women, with lower socioeconomic status (SES) are at an increased risk of hazardous alcohol consumption than those with higher SES.[2]
Alcohol control policies directed at reducing hazardous use of alcohol have been developed and implemented in several countries.[3] While these legislative interventions had considerable success in reducing the overall levels of hazardous alcohol consumption, their equity impact remains to be determined.
We aimed to, first, determine if a SES gradient in hazardous alcohol consumption exists in Geneva and, second, to evaluate the equity impact of the implemented alcohol control laws. We also sought to identify spatially-dependent geographical clusters of hazardous alcohol use that could benefit from targeted interventions
Methods
We used data from the Bus Santé study (n= for educational analyses and n= for occupation analyses), an ongoing population-based cross-sectional study in the State of Geneva, monitoring population’s health and its determinants.
Participants aged years were included. Abstinent participants were excluded (n=3059, 15.2%).
Educational attainment (divided into three categories - primary, secondary and tertiary) and professional occupation (divided according to the British Registrar General’s Scale into low, medium and high categories) were used as SES indicators.
We defined survey periods according to implemented alcohol control laws: period 1 (before 20/10/2000, baseline), period 2 (20/10/2000 to 01/02/ advertising ban), period 3 (02/02/2004 to 31/10/ % increase in alcopop price, decrease of alcohol driving limit, increased sales regulations) and period 4 (from 01/11/2009 onward – implementation of a public smoking ban). A consumption of >30g/day for men and >20g/day for women was defined as hazardous, according to the latest Swiss guidelines. We used Poisson regression models to determine the association between SES-indicators and the outcome variable.
We used the slope (SII, impact measure with similar interpretation to prevalence difference) and relative (RII, similar interpretation to prevalence ratio) regression-based indexes of inequality to quantify absolute and relative SES inequalities in hazardous alcohol use, respectively. These indexes were compared between periods with different alcohol control laws to assess their impact on SES inequalities.
We used the geographical coordinates of participants’ residence and the Getis-Ord Gi statistic to analyse spatial dependence of hazardous alcohol use and identify its hot spots in Geneva.
Results
1) Participants’ characteristics
2) The decrease in hazardous alcohol consumption has not occurred in men with lower education
*each shaded area corresponds to a legislative period
**proportions are age-adjusted
3) SES inequalities are observed in hazardous alcohol consumption, with a higher proportion of the outcome in men with lower SES, but not in women
indexes are adjusted for age, nationality and smoking status
Men
Women
Estimate (95%CI)
p value
Educational Attainment
Prevalence ratio:
primary vs tertiary
1.58 [1.39;1.80]
p<0.001
0.84 [0.70;1.00]
0.048
secondary vs tertiary
1.32 [1.18;1.47]
0.86 [0.74;0.99]
0.035
RII (least to most educated)
1.87 [1.57;2.22]
0.76 [0.60;0.97]
0.026
SII (least to most educated)
0.14 [0.11;0.17]
-0.04 [-0.07;-0.01]
0.008
Occupational level
low vs high
1.40 [1.24;1.59]
1.09 [0.81;1.45]
0.58
medium vs high
1.07 [0.93;1.24]
0.31
0.83 [0.70;1.00]
0.053
RII (low to high)
1.68 [1.38;2.06]
0.86 [0.62;1.20]
0.38
SII (low to high)
0.11 [0.07;0.15]
-0.02 [-0.05;0.02]
0.30
4) Alcohol control law packages had no effect on absolute (SII) and relative (RII) SES inequalities in hazardous alcohol consumption in men (red) and women (blue)
indexes are adjusted for age, nationality and smoking status. p< **0.01, ***0.001
same results were obtained with occupational level as SES indicator
5) Geographical hotspots of hazardous alcohol consumption were observed and could be the basis of targeted interventions
Conclusions
We identified the existence of a SES-related gradient in hazardous alcohol consumption in the State of Geneva. Furthermore, we showed that alcohol control legislative interventions could not obliterate SES-related inequalities. Additional interventions are still needed in order to close this SES-related gap, for instance, targeted interventions using the identified geographical clusters of hazardous alcohol consumption.
Overall
Men
Women
Primary education
Secondary Education
Tertiary education
p-value
N (%)
16725 (100%)
1257 (14.7%)
4119 (48.2%)
3173 (37.1%)
1750 (21.4%)
3414 (41.8%)
3012 (36.8%)
Age, mean ± SD
52.1 ± 10.6
52.8 ± 10.9
52.8 ± 10.7
51.0 ± 10.6
<0.001
54.6 ± 10.6
52.9 ± 10.4
49.8 ± 10.1
Swiss nationaltiy No
4704 (28.1%)
690 (54.9%)
964 (23.4%)
1054 (33.2%)
561 (32.1%)
568 (16.6%)
867 (28.8%)
Yes
12013 (71.9%)
567 (45.1%)
3152 (76.6%)
2116 (66.8%)
1189 (67.9%)
2846 (83.4%)
2143 (71.2%)
Hazardous alcohol consumption
0.62
13676 (81.8%)
840 (66.8%)
3089 (75.0%)
2641 (83.2%)
1510 (86.3%)
2979 (87.3%)
2617 (86.9%)
3049 (18.2%)
417 (33.2%)
1030 (25.0%)
532 (16.8%)
240 (13.7%)
435 (12.7%)
395 (13.1%)
Smoking status
Never smoker
6812 (42.5%)
379 (30.2%)
1356 (33.0%)
1403 (44.3%)
819 (53.1%)
1441 (46.2%)
1414 (50.1%)
Current smoker
3829 (23.9%)
382 (30.4%)
1154 (28.0%)
625 (19.7%)
355 (23.0%)
794 (25.4%)
519 (18.4%)
Ex-smoker
5385 (33.6%)
496 (39.5%)
1605 (39.0%)
1140 (36.0%)
368 (23.9%)
886 (28.4%)
890 (31.5%)
References 1) Global status report on alcohol and health, WHO ) Marmot M et al Lancet ) Tackling harmful alcohol use, OECD 2015