Contents lists available at ScienceDirect Psychiatry Research journal homepage: www.elsevier.com/locate/psychres A geographical exploration of environmental and land use characteristics of suicide in the greater Toronto area Eric Vaz a, ⁎ , Richard R. Shaker a , Michael D. Cusimano b a Department of Geography and Environmental Studies, Ryerson University, Toronto, ON, Canada b University of Toronto Dept. of Neurosurgery, St. Michael's Hospital, Toronto, ON, Canada ABSTRACT In developed countries, suicide has become one of the leading causes of mortality. With approximately 3500 cases taking place in Canada annually, it is currently the seventh-most common cause of death. A clearer spatial understanding of the suicide landscape in rapidly changing urban environments would especially help mitigate this problem. This study examines suicide rates in Toronto between 2004 and 2011 as to understand the spatial distribution of suicide by means of the importance of metropolitan places. The study uses geocomputation and statistical methods, enabling spatial analysis as tools to further assess the prevalent gender disparities of self-harm, advancing the current fndings in the suicide literature. The fndings clearly expose that the dichotomy of gender (male and female) produce diferent spatial patterns of self-harm, and are impacted by landscape characteristics diferently. Specifcally, the confguration of diferent land cover types have a much great impact on the female population than male. This spatial-exploratory understanding of not only the geographical distribution of rates, but also an assessment of landscape infuence can help to mitigate suicide depending on demographic and spatial-explicit characteristics found through advanced geostatistical and spatial analytical modeling. 1. Introduction Geographic Information Systems have become an essential corner- stone for epidemiology (Krieger, 2003; Cromley, 2019) as well as public health in recent years (Rushton, 2003; Kirby et al., 2017). The assess- ment and consequential understanding of policy and public health in- tegration with locational impacts on neighborhoods, as well as other clearly defned boundaries have allowed several health care sectors to take advantage of locational based analytics to gear towards smarter cities, sustainable development and designing adequate policies to cope with planning issues towards mental health issues (Bonnes and Bonaiuto, 2002). Several studies have assessed mental health within an epidemiological approach resorting to applied spatial analytical methods, advancing fndings directly in the feld of psychiatric re- search. Psychological distress, for instance, inferred by environmental and social determinants are strongly geographically clustered, and the prevalence of certain mental health issues can be assessed positively. The ``breeder hypothesis" and the ``drift hypothesis" are both, for in- stance, strongly linked to socio-economic characteristics which in- trinsically have confned and well defned spatial boundaries. The ex- tension of these issues to suicide, autism, and depression are but a few direct examples of the impact of psychiatric issues throughout geo- graphical space. The assertive locational characteristics brought by quantitative methods that spatially-explicit models allow to assess, remit the im- portance of these models within psychiatric research. This combination is possible due to the integrative approach of available geocomputa- tional data, as well as the signifcant advances in spatial methods in recent decades (Fig. 1). As seen in Fig. 1, while we can make a clear distinction in health data between spatial and non-spatial data they are signifcantly diferent. Spatial data due to its geographical character- istics intertwines new dimensions that bring relevant assets to (i) Psy- chiatric Epistemology, (ii) Mental Health Planning, (iii) Public Health Governance. Intrinsic to the discipline of Psychiatry, these distinct di- mensions can be combined utilizing spatio-temporal analysis together with geostatistics, allowing to understand dynamics of crucial in- dicators such as depression (Yang and Mu, 2015), suicide (Cheung et al., 2012a; Fontanella et al., 2018), self-harm, among others. Combining geostatistical analysis with landscape metrics, we may assess mental health planning structures, employing the physical distribution of pollution and chemical compounds that may have de- leterious efects on mental health (Helbich et al., 2012a), and whose research output may directly bring better local and regional planning (Helbich et al., 2017). The combination of such metrics when applied to land use and urbanization (Vaz et al., 2015), together with the ex- ploration of demographic indicators as allowed through spatial clus- tering techniques, enables spatial profling of zones and neighborhoods (Law and Perlman 2018), that allow psychiatric research to interact https://doi.org/10.1016/j.psychres.2020.112790 Received 13 July 2019; Received in revised form 9 January 2020; Accepted 12 January 2020 ⁎ Corresponding author. E-mail address: evaz@ryerson.ca (E. Vaz). Psychiatry Research 287 (2020) 112790 Available online 01 February 2020 0165-1781/ © 2020 Elsevier B.V. All rights reserved. T