BACKGROUND: Unintentional poisoning increased more than 500% in Oklahoma from 1999-2013, with 127 deaths in 1999 and 730 deaths in 2013. Of the more than 4,600 unintentional poisonings deaths from 2007-2013, 78% involved prescription drugs and 87% of those deaths involved opioids. While data are currently collected on fatal and hospitalized opioid poisonings in Oklahoma, a standard mechanism to capture nonfatal poisonings treated in emergency departments (ED) is not available. Recent research found that 91% of patients who had a nonfatal opioid poisoning were prescribed opioids after the event; 7% had a repeated opioid poisoning. Identifying nonfatal poisoning patients provides an opportunity to inform subsequent prescribing practices, treatment, and prevent future poisoning events.
METHODS: Oklahoma does not have a statewide ED discharge database. Surveillance was conducted to determine the magnitude of all patients treated for nonfatal poisoning in hospitals. Data for patients discharged in 2014 with an International Classification of Diseases, Ninth Revision, Clinical Modification primary diagnosis code of 960-979 were requested from 134 Oklahoma hospitals. Age, sex, admission and discharge dates, and discharge diagnoses of patients were obtained and entered into an Access database and demographic data were analyzed using SAS. Hospitals were selected for further medical record abstraction if any of the reported patients had a primary diagnosis code for opioids and related narcotics (965.00, 965.01, 965.02, or 965.09).
RESULTS: Ninety-three percent of hospitals responded for a total of 11,972 patient encounters for nonfatal poisoning. Of these, 1,539 patients (13%) were identified with a primary diagnosis of an opioid or other related narcotics poisoning, including 515 opium, unspecified (33%); 125 heroin (8%); 106 methadone (7%); and 793 other opiates and related narcotics (52%). 57% of patients were female and 50% were 30-59 years of age. 56 patients were treated in the ED or admitted to the hospital two or more times during 2014 with an opioid poisoning. Results from medical records abstractions of patients treated in the ED with an opioid poisoning will be presented.
CONCLUSIONS: This preliminary surveillance of statewide nonfatal poisonings provides a baseline for monitoring future trends. Of particular importance is heroin poisonings; deaths involving heroin have been increasing in the United States, but the magnitude is currently low in Oklahoma. Expanding surveillance to include fatal and nonfatal poisonings treated in hospitals will better inform prevention efforts and capture information on persons presenting to the ED at risk for a future poisoning event or fatal overdose.