Nearly 30 percent of deaths involving poisoning or toxic effects had at least one mention of the toxic effects of substances that were chiefly nonmedicinal. The toxic effects of alcohol and of carbon monoxide were more likely to be listed on death certificates than other toxic, nonmedicinal substances. About one-fourth of suicides involving poisoning and toxic effects had mention of carbon monoxide poisoning, and 9 percent of unintentional deaths involving poisoning and toxic effects included mention of alcohol.
Poisoning death rates increased with age from less than 1 per 100,000 for persons under 15 years of age to 19/100,000 at 35 to 44 years of age, and then declined again with age (Figure 3-2). For persons 15 to 19 and 20 to 24 years of age, death rates for males were about three times the rates for females; for those age 25 and older, the ratio was closer to 2:1 (data not shown).
The age-adjusted death rate for poisoning was 8.5/100,000, with rates ranging from a low of 1.8 for the Asian and Pacific Islander population to 9.8/100,000 for blacks. For each racial and ethnic group and both sexes, age-specific rates were higher for persons ages 35 to 44 and 45 to 54 than for those younger or older, with the highest rate for black males ages 45 to 54 (Figure 3-3).
Poisoning death rates are highest in the Mountain states (with New Mexico’s rate being the highest [16.4/100,000] in the United States) and lowest in the West North Central states (particularly North Dakota, South Dakota, Iowa, and Minnesota) and Pacific states (dominated by the low rate in California, 4.1/100,000). In each geographic division except for New England, the highest death rates were associated with unintentional poisoning, followed by poisoning by suicide. In New England, rates for poisoning of undetermined intent were higher than other poisoning death rates (Figure 3-4). The East North Central states had relatively high rates associated with psychoactive drug use mortality.
By combining the estimates of poisoning from the various data sources described, it is possible to develop a more complete estimate of poisoning incidence than can be obtained from any single source. It is interesting to note that while the distribution of estimated poisonings varies across regions for several of the data sources, these differences are blunted when all sources are combined. A comparison of annual episode estimates from the various data sources by level of care is presented in Table 3-11.
Both the NHIS and TESS provide estimates of poisoning events in which no direct (face-to-face) clinical evaluation or treatment occurred.