Dependence on alcohol and drugs can have deleterious effects on the psychological and physical well-being of individuals[1, 2]. Substance use has been associated with many adverse and long-term health outcomes including heart disease, liver failure, weakened immune system, lung disease, and mental confusion[3, 4]. Substance use also has psychosocial consequences such as breakdown of relationships and may lead to other risk behaviours (i.e. unprotected sex, etc.) by impairing decision-making and weakening the ability to process social cues[5, 6]. According to the 2016 global burden of disease report, alcohol use disorders are the most prevalent of all substance use disorders worldwide with an estimated age-standardised prevalence rate of 1320.8 per 100,000 population[7]. Marijuana and opioid dependence are the most common drug use disorders with an estimated 22.1 million cases and 26.8 million cases per year; and despite reductions in global smoking rates, tobacco use remains a leading risk factor for early death in more than 100 countries, accounting for 11.5% of global deaths in 2015 [7, 8].

A vast amount of research has documented risk and protective factors associated with substance use and misuse [1, 4, 9, 10]. While much of this research has focused on individual, peer, familial, and genetic correlates of substance misuse [11–14], a growing subset has focused on neighborhood environments as determinants of individual behaviors like substance use. Neighborhoods possess physical and social characteristics (e.g. crime, alcohol outlets, poverty, etc.,) that may impact individual health behaviors through various mechanisms, including physiological stress pathways [15]. Persistent exposure to noxious and threatening environments can induce a stress response which can impair health directly or indirectly through engagement in risk behaviors [16–18]. Substance use may be one method used to cope with social and environmental stressors such as neighborhood crime and violence [19].

Previous explorations of neighborhood-level determinants of substance use have primarily focused on indicators such as poverty, average household income, and economic disadvantage [20–24], as opposed to crime or safety specifically. Findings on the relation between these neighborhood exposures and substance use have been mixed, varying by type of substance and socioeconomic conditions [22, 23, 25–27]. A study by Galea et al., for example, reported that mean neighborhood income was associated with an increased likelihood of alcohol and marijuana use, but not cigarette use among a sample of adults in New York City [26]. Hill et al. reported a positive correlation between neighborhood disorder and heavy drinking [18], while Pollack and colleagues found higher rates of alcohol consumption in least deprived neighborhoods [28]. Inconsistent findings are due to many factors (i.e. study design, structural confounding, spatial misclassification, etc.,) including differences in how neighborhoods are operationalized.

While individual criminal behavior and substance use are highly correlated [29, 30], less is known about the relationship between neighborhood-level crime-related exposures and individual substance use [23, 31]. Crime and safety in one’s residential community, while highly correlated with poverty, may have a more proximal link to stress and health outcomes given their strong impact on physiologic systems. Exposure to violence has been found to influence the underlying neurobiological pathways related to threat perception, potentially altering an individual’s response to threat not only when exposed to violence, but also on a day-to-day basis [32]. In the case of neighborhood violence exposure, the perpetrator is not necessarily known nor the same person, triggering an adaptive constant vigilance response leading to a greater biologic toll.

Neighborhood crime and violence have been linked to several health outcomes across the lifespan, including obesity, hypertension, and adverse birth outcomes [33–37]. Studies have also positively linked subjective measures of neighborhood crime and lack of safety to smoking behaviors [38–43]. Research on the association of neighborhood crime-related exposures and alcohol and other drug use, however, is more limited [23, 26, 44, 45]. Also limited are studies that assess objective measures of neighborhood crime [41, 44, 46].

Many studies on neighborhood effects and health conceptualize place based on administrative boundaries (e.g. census tracts, zip codes) and aggregate measures of neighborhood characteristics (i.e. number of households below the poverty line, number of fast food outlets) (for a review, see Ross & Mirowsky, 2001) [16]. While defining neighborhoods in this way has advantages (i.e. standardized boundaries, quantification of area effects across a standard physical area, etc.,), this approach does not account for subjective perceptions of place, which may provide a different characterization of neighborhood social and physical environments than objective measures. Some researchers suggest that perceptions of one’s neighborhood may be even more important than objective, measurable factors in predicting health outcomes [47, 48]. Subjective interpretation of stressors plays an important role in determining the individual stress response [49]) and the ways in which individuals appraise and interpret neighborhood conditions. Results from studies using both objective and subjective measures of neighborhood characteristics have been mixed with varying levels of associations with substance use or other behavioral outcomes [46, 50–52]. The discordance among perceived and objective measures suggests that while they may be related, they are distinct concepts, which highlights the importance of using multiple methods to obtain a comprehensive understanding of crime and perceived safety.

In addition to perceived safety, a related measure—perceived neighborhood disorder—has also been linked to substance use behavior [21, 25, 53–56]. Neighborhoods with high levels of disorder have visible cues (i.e. graffiti, abandoned buildings, vandalism, crime, deviance, etc.,), which are often indicative of a breakdown in social control and weak community cohesion [57] and may evoke a physiological stress response [25] and therefore potential maladaptive coping behaviors [19].

The role of neighborhood crime and disorder on substance use may also differ by sex and gender. Research on crime and gender has shown that women have a greater fear of crime and perceive a greater risk of crime and victimization compared to men [58, 59]. Thus, it is likely that women and men perceive and experience neighborhood characteristics indicative of crime and safety differently which can influence their physiological stress responses and coping strategies. Results from studies that have investigated the role of gender as a potential effect modifier between neighborhood exposures and health outcomes have been mixed [26, 28, 60].While some evidence suggests that the health effect of residential context is larger for men, when specific environmental-level factors are examined residential context is related to health for both men and women. Further investigation of the differential health effects that contextual factors have on men and women is important for developing effective interventions.

The current study utilized data from the Jamaican National Drug Use Prevalence Survey to examine the role of both objective and subjective measures of neighborhood crime and disorder on alcohol, tobacco, and marijuana use among a nationally representative sample of Jamaican men and women aged 12–65 years. Compared to the United States, rates of drug and alcohol abuse in the Caribbean are relatively low [61], but recent epidemiological trends show a steady increase, especially in countries where the perceived risk is low and illicit substances are easily obtained [62]. However, data on substance use disorders in the Caribbean region is limited [3, 7, 62]. In 2016, as an effort to strengthen information on substance use among Caribbean populations, the National Council on Drug Abuse (NCDA) and the Organization of American States (OAS) in partnership with country governments launched an initiative to conduct drug use prevalence surveys in the region, the first survey (on which this study was based) was conducted in Jamaica [61].

The study adds to the scientific evidence base in several ways. First, the majority of studies investigating neighborhood effects have been conducted in the U.S., Canada, Europe. [23, 27, 63] The way in which neighborhoods are perceived and influence substance use behaviors in Jamaica likely differ from that of more developed western countries, especially considering recent changes (2015) to the country’s national drug act which loosened restrictions on personal use of marijuana [61]. Furthermore, crime remains a main public safety issue for Jamaicans and a significant threat to the country’s human and economic development [64]. Jamaica has homicide rates that are notably higher than both the regional and global averages falling within the top 10% globally with increases since 2015 [65, 66]. Second, the incorporation of perceived and objective markers of crime and disorder and their role on substance use behavior in a high crime context is also rare. Finally, contrary to many neighborhood studies that simply adjust for sex differences, we examine sex as a potential effect modifier.