In the early 19th century, even before the seminal work by Durkheim (Durkheim, 1951), there was a recognition that an individual’s suicide risk is not only determined by individual characteristics, but also influenced by the society within which they live (Goldney et al., 2008). Area factors associated with suicidal behaviour include social fragmentation, deprivation, and unemployment (Milner et al., 2013; Rehkopf and Buka, 2006). Smaller scale qualitative studies have also shown the importance of a person’s environment in contributing to their risk of suicidal behaviour, but these studies are limited in their generalisability (Dongre and Deshmukh, 2012; Widger, 2015b). The limitation of many of the previous quantitative studies (including Durkheim’s work) is that the data were either collected at an area or an individual level, and the two levels of data were not combined. These studies have been criticised for not being able to disentangle whether the area effect observed is due to the characteristics of the individuals living in these areas (compositional1 ), or because of characteristics of the area itself (contextual2 ), over and above that due to its composition. Statistical techniques are now available that can distinguish compositional or contextual effects allowing us to investigate whether individuals living in areas of concentrated poverty have a higher risk of suicide regardless of their own level of poverty. Multilevel modelling techniques allows the researcher to model individual relationships, group relationship, and the link between them. Studies have investigated the influence of compositional and contextual effects simultaneously on suicide risk in high income countries (HIC) (Agerbo et al., 2007; Collings et al., 2009; Denney et al., 2015; Hawton et al., 2001; Johnston et al., 2006; Maimon and Kuhl, 2008; Martikainen et al., 2004; Neeleman and Wessely, 1999; O’Reilly et al., 2008; Turnbull, 2014; Zammit et al., 2014), with few studies investigating attempted suicide risk (Hawton et al., 2001; Maimon and Kuhl, 2008). There is evidence from these studies that contextual factors influence suicide risk (Agerbo et al., 2007; Denney et al., 2015; Hawton et al., 2001; Martikainen et al., 2004; Zammit et al., 2014). To the best of our knowledge there have been no similar investigations in low and middle income countries (LMIC) (World Health Organisation, 2014). The present study investigates the effect of contextual factors on suicide attempt risk in a LMIC. The influence of contextual factors may be more pronounced in these settings because of the heavy emphasis on family relationships and collectivism (Bolz, 2002). In collectivist cultures the family unit and its characteristics form a strong part of a person’s identity and how they are viewed by other society members.