Trends in Death With Tuberculosis During the AIDS Era

Abstract

To describe and analyze recent changes in tuberculosis mortality in the United States during the acquired immunodeficiency syndrome (AIDS) epidemic. We used National Center for Health Statistics multiple-cause mortality data and analyzed deaths with tuberculosis (1980 through 1990) and/or AIDS (1987 through 1990) as an underlying or associated cause. We also categorized the 50 states and the District of Columbia into high (five states), medium (23 states), and low (23 states) AIDS incidence groups and then compared the groups' rates of death with tuberculosis during the period 1980 through 1990. Residents of the United States who died in the period 1980 through 1990. Death certificates indicating AIDS and/or tuberculosis. A bimodal age distribution of persons dying with tuberculosis has emerged concurrent with the AIDS epidemic. A new peak spanning the ages 20 to 49 years accompanies the preexisting peak in the elderly. In 1990, 54.2% (729/1344) of the deaths with tuberculosis in persons 20 to 49 years of age occurred in persons who also had AIDS listed on their death certificates. During the period 1987 through 1990, there was an increasing trend in the proportion of AIDS deaths that also had tuberculosis: 2.3% (353/15,075) in 1987, 2.5% (460/18,649) in 1988, 3.0% (738/24,607) in 1989, and 3.0% (836/27,975) in 1990 (P < .001). Of AIDS deaths in 1990, 1.6% (253/15,565) of whites died with tuberculosis, compared with 4.7% (400/8533) of blacks (P < .001) and 4.7% (172/3666) of Hispanics (P < .001). In the high AIDS incidence states, the rate of death with tuberculosis in persons aged 20 to 49 years rose from 0.61 to 2.82 per 100,000 population in the period 1982 through 1990, an increase of 362%. For this age group in the states with medium and low AIDS incidence, the rates of death with tuberculosis during this period rose 52% and 47% (from 0.44 to 0.67 and from 0.19 to 0.28 per 100,000 population), respectively. The increases in the rate of death with tuberculosis within all three groups and the differences among the groups were all highly statistically significant (P < .001). The AIDS epidemic has significantly increased the number and rate of tuberculosis deaths in younger adults through 1990, although in this study discrimination of deaths due to tuberculosis from deaths with tuberculosis was not possible. Vigorous efforts to prevent and treat tuberculosis and AIDS are indicated to reverse the trends we have described.