Volume 32, Number 6, November/December 2000
This article is also available in format
DIGESTS
Neighborhood Economic Conditions Influence AIDS Incidence in Massachusetts
Between 1988 and 1994, residents of the poorest and most densely populated Massachusetts neighborhoods had a markedly higher incidence of AIDS than those in the least-poor and least-dense communities; the differences amounted to more than 300 excess cases of AIDS per 100,000 residents in the disadvantaged neighborhoods. The incidence of AIDS varied widely by economic deprivation, race or ethnicity, and gender--from zero cases per 100,000 among white women in the least impoverished neighborhoods to 1,053 per 100,000 among black men in the most densely populated areas. These findings, from the first state-level analysis of the effects of economic inequality on the incidence of AIDS, underscore the importance to HIV prevention efforts of understanding the epidemic's dynamics vis-à-vis neighborhood economic resources.1
The analysts used the statewide AIDS surveillance registry to identify all cases of the disease reported for the period 1988-1994. Since these records lack information on individuals' socioeconomic status, the researchers examined 1990 census data to determine the economic characteristics of each AIDS patient's block group (a neighborhood unit with an average population of 1,000). Three economic measures were assessed: the proportion of residents living below the federal poverty line, population density (number of residents per square mile) and the proportion of households with an annual income of at least $150,000 (i.e., high-income households). On the basis of these data and 1990 census population estimates, the analysts calculated AIDS incidence rates, by neighborhood measures of economic well-being, for the state overall and separately for black, Hispanic and white men and women.
A total of 8,059 Massachusetts residents who received AIDS diagnoses in 1988- 1994 were classified by block group. The majority were men (81%) and were younger than 40 (68%). Sixty percent were white, and the rest were predominantly black (22%) or Hispanic (17%). More than half lived in neighborhoods where at least 10% of the residents were below the poverty line, population density exceeded 10,000 people per square mile and fewer than 2% of households had high incomes.
Statewide, the cumulative incidence of AIDS for the period was 128 cases per 100,000 persons. At the neighborhood level, the incidence climbed as economic deprivation increased and as population density rose. Neighborhoods where 40% or more of the population lived below the poverty line had an AIDS incidence (362 cases per 100,000) that was nearly seven times the incidence in communities where fewer than 2% of residents were below poverty (53 per 100,000). The incidence also differed by more than 300 cases per 100,000 between neighborhoods that housed 25,000 or more people per square mile (373 per 100,000) and those that contained fewer than 1,000 residents per square mile (40 per 100,000). Communities where at least 10% of households were high-income had less than half the AIDS incidence (69 per 100,000) of those in which fewer than 2% of households reached this income level (175 per 100,000).
Patterns of incidence for women and men in each racial or ethnic group generally mirrored the overall pattern. On every measure of neighborhood economic deprivation, white women had the lowest AIDS incidence. Rates increased from zero cases per 100,000 in the least-poor communities to 13 per 100,000 in the poorest and from six to 43 per 100,000 across the range of population densities; they fell from 34 to seven per 100,000 as the proportion of households with a high income increased. By contrast, among black women, the incidence of AIDS climbed from 133 to 442 cases per 100,000 as poverty increased, and rose from 183 to 403 per 100,000 as neighborhood populations became denser. It declined from 385 to 195 per 100,000 with increasing proportions of high-income households. Hispanic women's AIDS incidence grew from 131 to 352 cases per 100,000 with increasing poverty levels and dropped from 307 to 51 cases per 100,000 with rising proportions of high-income households. Displaying a distinctive pattern, the incidence among Hispanic women climbed rapidly from 150 cases per 100,000 in the least-dense neighborhoods to about 300 per 100,000 in areas of intermediate density, and fell back to 188 per 100,000 in the most densely populated communities.
The incidence of AIDS among white men rose steadily as neighborhood poverty increased (from 84 to 411 cases per 100,000) and as populations grew more dense (from 66 to 746 cases per 100,000); it fell from 196 to 120 cases per 100,000 with increasing proportions of high- income households. Among black men, as population density increased, AIDS incidence climbed steadily, from 335 cases per 100,000 to 1,053 per 100,000, the highest rate for any subgroup. Across levels of poverty, the incidence rose from 561 to 936 per 100,000 among black men, while increasing proportions of high-income households were associated with a drop in incidence from 807 to 782 cases per 100,000. Rates among Hispanic men ranged from 534 to 930 per 100,000 as neighborhood poverty increased, from 415 to 760 per 100,000 as population density increased and from 785 to 467 per 100,000 as high-income households increased.
Although AIDS risk is typically reported and examined in relation to sex and race or ethnicity, the analysts conclude that "these social categories are insufficient to describe the population burden of AIDS; data must additionally be stratified by measures of adverse living conditions." They add that understanding the effect of neighborhood economic resources on the risk of disease is critical, because "reducing the incidence of AIDS will depend vitally on approaches that promote the growth of social and economic resources in neighborhoods where AIDS is endemic."--D. Hollander
Reference
1. Zierler S et al., Economic deprivation and AIDS incidence in Massachusetts, American Journal of Public Health, 2000, 90(7):1064-1073.