Abstract

Suicide results from complex interactions between biological, psychological, and socioeconomic factors. At the population level, the study of suicide rates and their environmental and social determinants allows us to disentangle some of these complexities and provides support for policy design and preventive actions. In this study we aim to evaluate the associations between environmental and socioeconomic factors and demographi- cally stratified suicide rates on large temporal and spatial scales. Our dataset contains information about yearly suicides rates by sex and age from 2000 through 2017 along a 4000 km latitudinal gradient. We used zero- inflated negative binomial models to evaluate the spatio-temporal influence of each environmental and socio- economic variable on suicide rates at each sex/age combination. Overall, we found differential patterns of as- sociations between suicide rates and explanatory variables by age and sex. Suicide rates in men increases in middle and high latitude regions and intermediate age classes. For adolescent and adult women, we found a similar pattern with an increase in suicide rates at middle and high latitudes. Sex differences measured by the male/female suicide ratio shows a marked increase with age. We found that cloudiness has a positive effect on suicide rates in both men and women 24 years old or younger. Regional poverty shows a major impact on men in age classes above 35 years old, an effect that was absent in women. Alcohol and marijuana consumption showed no significant effect sizes. Our findings support high spatio-temporal variability in suicide rates in interaction with extrinsic factors. Several strong differential impacts of environmental and socioeconomic variables on suicide rates depending on sex and age were detected. These results suggest that the design of public policies and interventions to reduce suicide prevalence need to consider the local social and environmental contexts of target populations.