In the United States seven out of ten tuberculosis sufferers are minorities. During the last decade, the increase in the number of tuberculosis cases has been concentrated among minorities. The question motivating this research is why tuberculosis affects U.S. minorities disproportionately. Given current knowledge about the epidemiology of infectious diseases, risk factors for tuberculosis and the root of the decay of minority neighborhoods in this country, residential segregation seems to be one plausible answer to this question. The objective of this research is to elucidate the extent to which segregation has shaped the epidemiology of TB among U.S. minorities. This analysis is based on data for the state of New Jersey, where urban areas rank among the cities with the highest rates in the country. Individual level data on tuberculosis cases from 1985 to 1992 were combined with 1990 census socioeconomic information at the zip code level. Six segregation indices (dissimilarity index, isolation index and four measures of concentration) for zip codes, as well as measures of exposure to poverty, crowding and immigration for each racial and ethnic group were incorporated in the analysis. A novel Boolean-logic analytic methodology was used, in order to identify configurations of risk factors and segregation dimensions which distinguish communities where TB is very high from three other types of community (high TB, low TB and no TB). The main findings are: (i) A combination of risk factors (poverty, crowding, dilapidation, immigration, proximity to New York City and high population densities) operates in zip codes with very high TB, and significantly distinguishes this type of community from the rest; (ii) For African Americans and all minority groups combined, segregation along two dimensions, i.e. isolation and concentration, significantly differentiates very high TB communities; (iii) The interaction of segregation (i.e. high isolation levels, often coupled with a high density of black individuals), poverty and crowding has shaped the epidemiology of tuberculosis among African Americans; (iv) Immigration (in combination with poverty and crowding) seems to play a more important role in the epidemiology of TB among hispanics and Asians than among blacks.