580,000 released records (United States Pharmacopoeia, 2002). This program’s relevance to poisoning and drug overdose surveillance systems is limited given that it is a voluntary, subscription-based method for reporting (no public access for analysis), weighted heavily to therapeutic misadventures.
The Hazardous Substances Emergency Events Surveillance system is a state-based program managed by the Agency for Toxic Substances and Disease Registry. As of 2002, the ATSDR compiled data detailing hazardous substance emergencies derived from 15 states. Although the primary unit of analysis in this system is the release event, details on the number of persons affected and associated health care utilization are obtained. The data in this system are most relevant to airborne releases, particularly of irritant gases (Berkowitz et al., 2003; Horton et al., 2002; Orr et al., 2001; Weisskopf et al., 2003).
A specialized surveillance dataset of potential interest is the Census of Fatal Occupational Injuries, maintained by the federal Bureau of Labor Statistics (BLS) (Valent et al., 2002). The goal of this system is to capture all occupationally related fatalities in the United States, including cases resulting from toxic exposures. This dataset is notable in that it includes data from all 50 states and attempts to integrate and cross-check data from multiple sources, including death certificate data, workers’ compensation insurance claims, Occupational Safety and Health Administration (OSHA) fatality reports, and media reports. BLS also oversees two national surveys for nonfatal occupational injuries; one survey is based on a sample of workers’ compensation claims and the other on OSHA-mandated injury forms. Neither system is effective in detecting toxin-related events; thus, each is marginal in relation to poisoning surveillance generally.
In addition to data derived from poisoning and drug exposure sources and from general vital statistics and medical encounters, national survey data are also relevant to poisoning surveillance. Of these, the most comprehensive is the annual National Health Interview Survey (NHIS). This survey, based on an extensive national stratified random sample, generates data on poisoning through a series of injury-related items asked of all survey recipients. Beginning in 2000, coverage was expanded to better capture poisoning events. Items address the general nature of the exposure as well as some details of circumstances involved, including whether a poison control center was contacted. Only limited analyses of poisoning data derived from NHIS have been published by independent investigators (Fleming et al., 2003; Polivka et al., 2002).