Macrosocial determinants of risks for alcohol-related trauma among lower income populations

Abstract

Alcohol-related injuries place a substantial burden on public health, and contribute to health disparities between higher and lower income populations. This thesis assesses a set of theoretical causal mechanisms that combine to produce more problems in lower income areas. The mechanisms are described in two hypotheses, (i) that the economic geographic processes common to all retail markets will produce greater concentrations of alcohol outlets in lower income areas; and, (ii) that alcohol-related trauma will occur with greater frequency in areas with greater concentrations of alcohol outlets. I test each hypothesis individually, then simulate the combined effects of this “causal in a mathematical simulation. The first hypothesis is based on theory from economic geography and urban economics describing the spatial dynamics of retail markets. Because higher income populations consume more alcohol than lower income populations, and alcohol outlets are for-profit business that seek to minimise costs and maximise profits, outlets will be attracted to higher income areas. However, outlets will also be excluded from these areas due to greater land and structure rents, so they will appear in greater concentration in lower income areas proximate to higher income areas. In a series of three papers I demonstrate that geographic distributions of alcohol outlets are consistent with these theoretical predictions (i) within a single metropolitan area in Australia, (ii) between towns in a rural area of Australia, and (iii) across the extent of cities within one U.S. state. Regarding the second hypothesis, numerous studies have demonstrated that areas with greater concentrations of on-premise alcohol outlets (e.g., bars) have greater incidence of traumatic injuries, and that these relationships are observable over time and in a range of geographic settings. The literature regarding off-premise outlets (e.g., liquor stores) is less developed, but is suggestive that similar relationships may obtain. Based on these previous studies and theory regarding economies of scale within outlets, and access to alcohol within populations (i.e., availability theory), I propose that areas with greater concentrations of off-premise outlets will have greater incidence of traumatic injuries, that larger and chain outlets will sell cheaper alcohol than smaller and independent outlets, and that larger and chain outlets will contribute most substantially to trauma risk. I assess these relationships using a combination of archival data (i.e., Census, planning and zoning, and ambulance event counts) and originally collected data from field observations in a sample of off-premise outlets. Results support the proposed mechanism. Finally, I test the proposed causal architecture using a mathematical simulation. I construct three synthetic cities containing uniform geographic distributions of population and income. Using point estimates extracted from the previous studies, I then estimate the number of alcohol outlets within the spatial units that comprised the cities, then the number of traumatic injuries due to alcohol outlets within each spatial unit. Reducing the income for one spatial unit produces greater incidence of traumatic injury in that unit, but greater incidence elsewhere (i.e., in adjacent units, in other units within the same city, and in the units of other cities). Increasing the income for one spatial unit has the reverse effect. The simulation provides evidence that a broad causal architecture contributes to disparities in health between higher and lower income populations due to alcohol-related trauma. Crucially, these processes are an inevitable product of the spatial dynamics common to most retail markets. Interventions that disrupt the market forces which produce greater concentrations of outlets within lower income areas (e.g., density limits, distance limits between outlets) may limit the extent to which lower income populations are disproportionately exposed to risks.