Dissemination of information to the public and clinicians is the primary role of poison information centers. The information is provided by specialists in poison information (nurses and pharmacists) and clinical toxicologists. The poison control specialists document each patient interaction electronically; these interactions form the basis for the poison center medical record. Additionally, if a physician treating a patient calls the center, he or she records the center's advice on the patient's medical record in the clinical office or emergency department.
The poison control center's record contains the standard personal and demographic information about the patient as well as specific information that are categorized by substance (the poison), treatment, patient symptoms, route of exposure, and laboratory values. Each of these areas constitutes a fully searchable section of the medical record. The values within each section are standardized and coded so that data from all poison centers in the United States can be incorporated into a single database. There is also a free-text documentation section. The free-text section is not searchable as part of the standard poison center medical record, and is therefore not included in the national database on poisoning exposures.
Currently, participating poison control centers submit data to the American Association of Poison Control Centers Toxic Exposure Surveillance System (AAPCC TESS) on a monthly, quarterly, or semiannual basis. Sixty-four poison centers submitted data in 2003 and these data reflect all 50 states and the Commonwealth of Puerto Rico (Watson et al., 2003). Participation in AAPCC TESS and submission of all data is mandatory for certified regional poison information centers and voluntary for noncertified centers. Data are reviewed and published on an annual basis approximately nine months after the end of the calendar year. Since data submission is intermittent and partially voluntary, there is no real-time surveillance of national poison center data for toxidromes that are consistent with exposure to biological and chemical terrorism agents. Furthermore, most individual centers do not conduct real-time toxico-surveillance. Figure 25.3 shows the type of surveillance that is feasible using poison center data.
However, poison information centers collect a large volume of information on exposures involving both humans and animals. For example, the Pittsburgh Poison Center responded to
approximately 78,000 inquiries in 2000. Nearly 5000 of the inquiries involved animal exposures. This reflects the public awareness of poison information services. This call volume, the active data reservoir, and the reliance upon the poison center by the public and medical professionals make the poison center an ideal active surveillance center to identify sentinel events and to profile medically and demographically the nature of a biological or chemical terrorist event.
Software to facilitate the identification of sentinel events or toxidromes through the surveillance of real-time poison center data has not yet been widely deployed, though at least one vendor offers software which claims to tap into these data (Stout Solutions, LLC, 2005a). Currently, the only way to conduct analysis of data is through the identification of tox-idromes based on expert opinion and then the tedious analysis of data to identify the presence of those toxidromes. The data are available readily and are accessible easily through the use of any database management system. Most poison centers document cases in real-time, which make the data fields available for immediate analysis. It would not be a challenging task to conduct real-time background surveillance of a single poison center's medical record database. However, poison centers within most states and throughout the nation are not linked in real-time and all national data analysis is conducted retrospectively. The challenge is to network U.S. poison centers so that data are contributed in real time. Artificial intelligence may help to identify sentinel events that are occurring, below the established threshold of recognition.The collaboration of data that includes poison center data, nonprescription and prescription pharmaceutical sales, emergency department diagnoses, veterinary clinic data on companion and large animal problems, and so on could help to identify problems in the early stages of their evolution.
Was this article helpful?