Research paperAssociation between neighbourhood socioeconomic characteristics and high-risk injection behaviour amongst injection drug users living in inner and other city areas in Montréal, Canada
Introduction
Two decades of risk factor research has yielded a comprehensive understanding of the individual-level determinants of HIV and viral hepatitis C (HCV) transmission amongst injection drug users (IDUs) (Backmund, Reimer, Meyer, Gerlach, & Zachoval, 2005; Patrick et al., 1997, Santibanez et al., 2006). A high burden of viral outcomes and high rates of HIV and HCV transmission persist in many urban settings despite ongoing virus reduction efforts (ASSSM, 2006, Vancouver Coastal Health News Release, 2003). The limited utility of actions to curtail these viral epidemics by promoting individual behaviour change suggests a need to consider contextual factors that might positively or negatively condition high-risk injection behaviour (HRIB) and blood–borne transmission of HIV and HCV. A first step in assessing any such conditioning of health-related behaviour is to evaluate heterogeneity in place-based risk conditions in relation to group- and area-based differentials in risk factors and their health consequences (Daniel, Moore, & Kestens, 2008).
Several studies have shown that blood–borne virus infections and HRIB among IDUs are not distributed homogeneously within city boundaries (Brugal et al., 2003, Diaz et al., 2001; Hutchinson, Taylor, Goldberg, & Gruer, 2000; Maas et al., 2007, Miller et al., 2004; Rockwell, Deren, Goldstein, Friedman, & Des Jarlais, 2002; Wood et al., 2002). In comparison with their counterparts residing in less central surrounding areas, IDUs in the inner city of Barcelona, Spain, were more likely to be infected with HIV (Roca et al., 1995). In contrast, IDUs within inner city areas of Sydney, Australia had a lower likelihood of practicing syringe sharing relative to those in the surrounding areas (Darke, Hall, & Swift, 1994). These incongruent observations may relate to differences in local contextual conditions, specifically, the potential influence of small-scale local neighbourhood factors embedded within larger-scale areas (e.g., inner city vs. surrounding city areas) on HRIB.
Attempts to frame the influence of neighbourhood characteristics on risk of viral acquisition among IDUs can be guided by the conceptual framework for cities and population health developed by Galea, Freudenberg, & Vlahov (2005). This framework theoretically situates population health outcomes in an ecological system involving multiple levels of influence that condition health and specific determinants of health. It seeks to broadly account for the influence of higher levels of context (e.g., policies and regulations, services provision, etc.) on local living conditions relevant to various health outcomes. For IDUs and IDU outcomes, an inner vs. outer city distinction may arguably differentiate a higher order of context within which local neighbourhood conditions could vary in their associations with HRIB. Collective features of inner city areas as these typify many urban environments may either harm or promote health, and these impacts could well be different from those that prevail in surrounding areas outside of the central urban core. On the one hand, inner cities provide easier access to illicit drug markets and high-risk injection networks, and are characterised by undesirable environmental conditions including poverty, violence, deteriorated built environments, and considerable income and education disparities (Séguin & Divay, 2002; Wasylenki, 2001). On the other hand, inner cities can sometimes provide easier access to social and health services, particularly for marginal sub-populations such as IDUs.
Income and educational attainment aggregated at the level of small administrative units have been emphasised as two key features of local environments that may influence population health and health-related behaviour in urban areas (Ompad, Galea, Caiaffa, & Vlahov, 2007). In American cities, individual injection drug use patterns were found to be associated with living in an economically disadvantaged neighbourhood and/or a low educational attainment neighbourhood (Bluthenthal et al., 2007; Buchanan, Shaw, Teng, Hiser, & Singer, 2003; Fuller et al., 2005; Galea, Ahern, & Vlahov, 2003). Other research indicates that neighbourhood income is more strongly related to the health of low-income residents than that of the more advantaged (Boardman, Finch, Ellison, Williams, & Jackson, 2001; Stafford & Marmot, 2003). It is unclear however whether neighbourhood income and education have a unique influence or if the influence of one dimension is spurious due to association with the other (Galea & Ahern, 2005; Wen, Browning, & Cagney, 2003).
The measurement of local neighbourhood attributes relevant to health outcomes is a challenging task. In a recent special issue of the journal, Cooper et al. applied geographic information system (GIS) techniques, specifically buffer zones, to create geographic measures of syringe exchange program access and law enforcement activities at small geographic area levels (Cooper, Bossak, Tempalski, Des Jarlais, & Friedman, 2009). Likewise, the use of GIS to create circular buffers centred on residents’ homes has been used elsewhere to study the associations between local contextual conditions and health outcome (Chaix, Merlo, Subramanian, Lynch, & Chauvin, 2005; Chaix, Merlo, & Chauvin, 2005; Chaix et al., 2006). Buffers can be used to obtain summary measures of local factors which can then be examined in relation to the health outcome of interest, with the size of these circular surfaces (e.g., a given radius around each residential address) being based on the study's purpose (Berke, Koepsell, Moudon, Hoskins, & Larson, 2007; Crawford et al., 2008, Pate et al., 2008). For IDUs as well as the average resident, a 10-min walking distance, corresponding to a 500-m buffer, is generally recognised as a reasonable radius by which to represent access to local services (Cooper et al., 2009; Rockwell, Des Jarlais, Friedman, Perlis, & Paone, 1999; Rockwell et al., 2002).
Small circular buffers centred on an individual's residential address may be a more meaningful way to represent immediate residential environments, rather than large groupings of people within the artificial boundaries of administrative units. Administrative units such as census tracts, often used as proxies for neighbourhoods, are assumed to be homogeneous with respect to population characteristics, socioeconomic status and living conditions (US Census Bureau, 2007). Such representations however may not coincide with health-related processes (Diez-Roux, 2001, Diez-Roux, 2007).
In this study, we hypothesised that for IDUs on the Island of Montréal HRIB would be (1) more prevalent in inner city areas, (2) associated with neighbourhood socioeconomic living conditions including educational attainment and low income, and (3) differentially associated with neighbourhood conditions according to inner-city vs. surrounding area residence.
Section snippets
Setting
This study was conducted on the Island of Montréal, with a population 1.8 million residents and a land base of 500 km2 divided into 27 boroughs. Each borough is locally responsible for governance and municipal services. Ville-Marie borough is widely regarded as the inner city of Montréal. This central and oldest portion of the city is characterised by the highest levels of socioeconomic disadvantage and crime of all boroughs in Montréal (Savoie, Bédard, & Collins, 2006; Service de la mise en
Results
Of the 468 IDUs included in this study, a majority (84%) was male, with mean age (standard deviation) 40 years (±9), women being younger (33 ± 10). Nearly half of study participants resided within inner-city areas. Thirty-five percent of the sample engaged in HRIB.
Table 1 contrasts sociodemographic measures, drug injection patterns and health status between IDUs living within the inner city and IDUs from other boroughs. Rates of HRIB were similar for both groups. IDUs from inner city areas were
Discussion
The alternate hypothesis that a higher proportion of IDUs in inner city areas would report HRIB compared to IDUs in other city areas could not be accepted: similar proportions of HRIB were observed for both groups. On the other hand, HRIB was differentially related to neighbourhood socioeconomic characteristics between inner city and surrounding areas: associations were observed for neighbourhood educational attainment and low income for IDUs in the inner city, while relationships were null for
Conflict of interest
All authors declare no competing interests, including no financial, personal or other relationships with people or organisations within 3 years of beginning the work that could inappropriately influence, or be perceived to inappropriately influence, the work.
Acknowledgments
We thank Karine Léger for her assistance in merging Geographic Information System and cohort databases, and the staff and participants of the St. Luc Cohort.
Funding: This study was supported by a project grant from the Canadian Institutes of Health Research (#135260, JB and MD), with additional support from the Canada Foundation for Innovation (#201252, MD), and the Réseau SIDA et Maladies Infectieuses du Fonds de la Recherche en Santé du Québec (JB). Dr. Bruneau holds a clinical research
References (54)
- et al.
Geographic approaches to quantifying the risk environment: A focus on syringe exchange program site access and drug-related law enforcement activities
The International Journal of Drug Policy
(2009) - et al.
Framing the biosocial pathways underlying associations between place and cardiometabolic disease
Health & Place
(2008) Neighbourhoods and health: Where are we and where do we go from here?
Revue d’Epidemiologie et de Sante Publique
(2007)- et al.
Cities and population health
Social Science & Medicine
(2005) - et al.
Social epidemiology and complex system dynamic modelling a applied to health behaviour and drug research
The International Journal of Drug Policy
(2009) - et al.
Neighborhood and HIV infection among IDU: Place of residence independently predicts HIV infection among a cohort of injection drug users
Health & Place
(2007) - et al.
Physical activity and neighbourhood resources in high school girls
American Journal of Preventive Medicine
(2008) - et al.
Poverty, affluence, and income inequality: Neighbourhood economic structure and its implications for health
Social Science & Medicine
(2003) - ASSSM (Agence de la santé et des services sociaux de Montréal). (2006). Recrudescence d’infections par le virus de...
- et al.
Hepatitis C virus infection and injection drug users: prevention, risk factors, and treatment
Clinical Infectious Diseases
(2005)