Search Results

NACCT Abstracts   
Death and Liver Injury Following Repeated Acetaminophen (APAP) Ingestions by Children

Heard KJ,1 Bond RF,2 Clark RF,3 Dart RC,1 Green JL,1 Koff RC,4 Kozer E,5 Mlynarchek S,1 Steitz AG.1

1Rocky Mountain Poison & Drug Center, Denver Health, Denver, CO, USA; 2Department of Emergency Medicine, Cincinnati Childrens Hospital, Cincinnati, OH, USA; 3Department of Emergency Medicine, University of California San Diego, San Diego, CA, USA; 4Department of Medicine, University of Connecticut School of Medicine, Farmington, CT, USA; 5Pediatric Emergency Unit, Assaf Harofeh Medical Center, Zerifin, Israel

Background: While death from therapeutic APAP doses has not been reported in prospective studies there are retrospective reports describing fatalities following APAP administration with therapeutic intent, but not necessarily at therapeutic doses. This has resulted in confusion related to the safety of APAP in children. We characterized cases of clinically significant hepatic events following administration of multiple APAP doses. Methods: Retrospective cohort study of children <6 years with ALT >100 IU/L or death following >1 APAP dose. We excluded cases with insufficient information to determine dose and outcome (serum ALT, reported liver failure or death). Data sources were NPDS, FDA/AERS, medical literature and manufacturer internal safety reports. Cases were abstracted and reviewed by a five member expert panel to determine relationship of the hepatic event or death to APAP, the estimated APAP dose based on the reported dose and assessment of objective support for the dosing history (e.g. serum APAP concentration). Results: A total of 2,531 cases were reviewed and 146 unique cases met inclusion criteria with sufficient information. One hundred and two (70%) were rated as at least potentially related to APAP; 60 (40%) were <1 year and 37 (36%) were male. There were 26 deaths; 10 (38%) were <1 year and 9 (35%) male. Dose was therapeutic (<75 mg/kg/day) in 6 (6%) cases. Age range for therapeutic cases was 3 months to 4.5 years. The reported range of therapeutic doses was 23 mg/kg/day × 2 days to 60 mg/kg/day × 11 days. The lowest fatal dose where the dose history was consistent with other clinical information was 100 mg/kg/ day. Conclusions: This study was limited to reported cases with sufficient information and dosing information and may be subject to recall bias. Given the vast experience with APAP in children, reported ALT elevation in children given doses of <75 mg/kg/day of APAP is an extremely rare event. No deaths were attributed to therapeutic APAP doses. While our methodology may fail to detect some cases with asymptomatic ALT elevation, it is unlikely to miss deaths. Safety efforts should be directed at preventing inadvertent APAP overdose in children.

 Has One Pill Killed? A Review of AAPCC Pediatric Fatality Data

Monte AA, Yin S, Heard KJ, Bronstein AC.

Rocky Mountain Poison & Drug Center, Denver Health, Denver, CO, USA

Background: According to the 2009 American Association of Poison Control Centers (AAPCC) annual report, 51.8% of all exposures are in children <6 years of age.

Most common pediatric fatalities 1999–2008
Agent Total (%)
Prescription opioidsa 21
Carbon monoxide 17
Hydrocarbons 7.5
Analgesicsa 5.6
Cardiovasculara 4.9
Anticholinergics 4.5
aOn composite list.  

Several authors have proposed lists of substances that could be fatal in children at low doses. These lists have not been systematically validated and largely ignore non-pharmaceutical agents. Our aim was to determine which poisons have resulted in deaths in children <6 years of age reported by AAPCC. Methods: AAPCC annual reports from 1999 to 2008 were reviewed and all fatalities in children <6 years of age were included in analysis. Age, poison, route of exposure and intent of exposure were abstracted. When more than one class of poison was reported, the published death abstract was reviewed to determine the substance most likely responsible for the death. Agents responsible for deaths were cross-referenced with a composite list of 11 poisons (tricyclic antidepressants, antipsychotics, quinine/quinidine, calcium channel blockers, opioids, oral hypoglycemics, camphor, clonidine/imidazolines, diphenoxylate/ atropine, salicylates, and toxic alcohols) gathered from three review articles of agents theoretically able to kill in low doses. Results: A total of 267 fatalities were included. The median age was 1.8 years (IQR 1,3). Most (63%) of the fatalities involved pharmaceutical exposures (see Table). Agents on the composite list of deadly poisons were associated with 29% of the deaths. Quinine/quinidine, camphor and oral hypoglycemics were not responsible for any deaths. Hydrocarbons, analgesics and anticholinergics were common, but noticeably absent from the composite list. The most common reasons for exposure were unintentional (65%), therapeutic error (17%) and malicious (13%). Conclusions: Prescription opioids and carbon monoxide were the most common reported fatal exposures in children <6 years of age. While it is still important for physicians to recognize which agents can be harmful in children at low doses, those agents are not responsible for the majority of deaths reported to poison centers.

 Pediatric Chewing Tobacco & Snuff Exposures: 2000–2009

Deutsch CM,1 Bronstein AC,1 Banerji S,1 Hesse CL,1 Spyker DA.2

1Rocky Mountain Poison & Drug Center, Denver Health, Denver, CO, USA; 2Uniformed Services, University of the Health Sciences, Bethesda, MD, USA

Background: Tobacco exposures in the pediatric population pose a poison center (PC) management dilemma. These highly accessible adult products often come in attractive non-child resistant packaging. Little data is available on outcomes of children exposed to chewing tobacco or snuff. We sought to characterize the frequency of these exposures and the associated morbidity. Methods: We accessed National Poison Data System (NPDS) aggregate data human single substance exposures by year from 2000 to 2009 using AAPCC Generic Codes 0201057 (chewing tobacco) and 0201058 (snuff). Case counts for all human exposures and for children <6 years were obtained. Further descriptive data for children <6 years were examined (route, reason, healthcare facility management and medical outcomes). Calls over time for each measure were examined using regression (versus time) for linear and logarithmic (proportional) models to calculate % increase/year and 95% confidence intervals using SAS JMP v 6.0.0. Results: Ten thousand eight hundred and thirteen human exposures were identified and 9,610 (88.9%) of these were children <6 years old. Ingestion was the most common exposure route for both substances in all years (99.2%). Almost all exposures (99.8%) were unintentional. Over a quarter of cases (26.9%) were managed in a healthcare facility. There were no deaths. The most common reported outcome was “No Effect” (34.9%). Chewing tobacco exposures are increasing at >5%/year and accounts for most of the 3.48%/year increase in the total. The number of exposures in 2009 were 785 for chewing tobacco and 423 for snuff. Conclusion: Ten years of NPDS chewing tobacco and snuff exposure data revealed these nicotine products as a predominately pediatric exposure phenomenon. No deaths and very few major outcomes were observed. PCs may use this data to conclude that home management is usually appropriate. Further study is indicated to determine dose response.

Table. Tobacco exposures for children <6 years old from 2000 to 2009
Outcome Increase [95% confidence interval](%/year)
  No effect Minor Moderate Major Death Other Total cases
Chewing tobacco 2,065 1,911 141 2 0 1,880 5,999 5.05 [1.97, 8.13]
Snuff 1,288 1,145 91 2 0 1,085 3,611 0.895 [-1.71, 3.50]
Totals 3,353 3,056 232 4 0 2,965 9,610 3.48 [1.78, 5.18]


 Fomepizole for Severe Disulfiram-Ethanol Reactions

Sande MK,1 Thompson DK,1 Schaeffer TH,2 Monte AA.2

1Department of Emergency Medicine, Denver Health, Denver, CO, USA; 2Rocky Mountain Poison & Drug Center, Denver Health, Denver, CO, USA

Background: When ethanol (EtOH) is consumed with disulfiram (DSM), acetaldehyde accumulates and an unpleasant histamine-like reaction occurs. Severe DSM-EtOH reactions result in hypotension, tachycardia and angioedema. Fomepizole (4-MP), an inhibitor of alcohol dehydrogenase, may halt progression of this reaction by blocking EtOH metabolism to acetaldehyde. Methods: We present two cases of DSM and EtOH overdose leading to severe reactions unresponsive to fluid resuscitation and treated with 4-MP. Serial blood ethanol concentrations (BAC) were used to determine kinetics of EtOH elimination following 4-MP blockade. Results: Case 1: A 20-year-old female presented after ingestion of vodka and 7,500 mg DSM. Presenting HR was 125 bpm and BP 119/83 mmHg. BAC was 448 mg/dL, drugs of abuse screen and laboratory tests were otherwise normal. After 11 h of observation and 2 L of normal saline, she developed skin flushing, lip swelling, tachycardia (166 bpm) and hypotension; systolic blood pressure (SBP) 88 mmHg. Antihistamines, steroids and an additional 2 L of normal saline were given without improvement of hypotension or tachycardia. One dose of 4-MP 15 mg/kg was given with improvement within 1.5 h; BP 117/44 mmHg and HR 98 bpm. No additional doses of 4-MP were given and there was no recurrence of tachycardia or hypotension. Mental status normalized over 16 h and she was discharged with no clinical sequelae. Serial BACs demonstrated first order elimination kinetics, even after blockade. Case 2: A 47-year-old female presented after overdose of vodka and 6,250 mg DSM. She was tachycardic and hypotensive upon presentation. After administration of 3 L normal saline, she remained hypotensive and tachycardic. BAC was 221 mg/dL, drug of abuse screen and laboratory tests were within normal limits. One dose of 4-MP 15 mg/kg was given with improvement within 1 h: blood pressure and heart rate normalized. There was no recurrence of hypotension and her mental status cleared over 12 h. As in Case 1, BACs declined rapidly and appeared to follow first order elimination kinetics. Conclusions: 4-MP is effective for the treatment of severe DSM-EtOH reactions. Analysis of BACs following blockade with 4-MP in both cases demonstrated first order elimination kinetics.

 Low-Molecular-Weight-Heparin Overdose: Intervention or Observation? 

Monte AA, Bodmer M, Schaeffer TH.

Rocky Mountain Poison & Drug Center, Denver Health, Denver, CO, USA

Background: Low-molecular-weight-heparins (LMWH) are commonly used in both inpatient and outpatient settings. Overdose of LMWH has rarely been reported and optimal treatment has not been established. Methods: We present three episodes of LMWH overdose in two patients and discuss clinical presentation, laboratory evaluation, therapeutic options, and patient outcomes. Results: Case 1: A 35-year-old female with systemic lupus erythematosus complicated by antiphospholipid syndrome and multiple thromboembolic events injected 72,000 units of dalteparin. Anti-Xa activity peaked at 6.2 U/mL (therapeutic 0.6–1.1 U/mL) 7.5 h post-injection and returned to the therapeutic range within 20 h. The patient had a similar overdose 20 days later, again 72,000 units of dalteparin was injected. Anti-Xa activity was 4.5 U/mL 2 h post-injection. The patient was discharged without clinical effects on both occasions. Case 2: A 29-year-old male with past medical history of antiphospholipid syndrome and pulmonary emboli presented after injection of 480 mg of enoxaparin. Anti-Xa activity 2 h post-injection was 1.9 U/ mL and decreased to 1.4 U/mL 14 h post-injection. The patient was discharged without clinical effects from the overdose. Neither patient received agents to reverse anticoagulation and no clinical bleeding was observed. Discussion: Literature regarding the efficacy of LMWH reversal agents was reviewed and revealed that protamine is inefficient at neutralizing LMWH. High dose protamine: LMWH (5:1 mass ratio) neutralized only 54.2% of anti-Xa activity for enoxaparin and 74% for dalteparin. An in-vitro study found that recombinant activated Factor VIIa (rFVIIa) effectively reversed anticoagulation effects of enoxaparin. Conclusions: Given the benign clinical course in these three reported episodes, withholding aggressive interventions, such as blood product transfusion, protamine administration, or hemodialysis may be advisable in absence of clinically relevant bleeding. However, in patients with clinically significant bleeding, use of protamine sulfate is reasonable and a trial of recombinant factor VIIa (rFVIIa) may be warranted for refractory cases.

Outcomes of Acute Acetaminophen Overdose Patients Treated with Less than 20 Hours of Intravenous N-Acetylcysteine

Yarema M,1 Lucyk S,2 Johnson D,1 Sivilotti M,3 Nettel-Aguirre A,1 Victorino C,1 Bailey B,4 Spyker D,5 Dart R,6 Rumack B.7

1University of Calgary, Calgary, AB, Canada; 2University of Alberta, Edmonton, AB, Canada; 3Queens University, Kingston, ON, Canada; 4Universite de Montreal, Montreal, QC, Canada; 5University of California-San Francisco, San Francisco, CA, USA; 6Rocky Mountain Poison and Drug Center, Denver, CO, USA; 7School of Medicine, University of Colorado Denver, Denver, CO, USA

Background: While the intravenous N-acetylcysteine (IV NAC) protocol for acute acetaminophen (APAP) overdose mandates at least 20.25–21 h of therapy, not all patients receive the full infusion, possibly due to risk reassessment, adverse reactions or physician preference. This practice is not widely studied, and often against poison centre advice, so we sought to describe outcomes for patients treated with fewer than 20 h of IV NAC following acute APAP poisoning. Methods: We utilized data collected as part of a structured medical record review of patients hospitalized for APAP poisoning at one of 34 Canadian hospitals from 1980 to 2005. We selected patients with a history of an acute (over less than an 8-h period), single ingestion with a potentially toxic serum APAP concentration (>150 μg/mL at 4 h) measured between 4 and 24 h postingestion who were started on the 20-h IV NAC protocol within 24 h of ingestion but received less than 20 h of IV NAC. Patients in whom aminotransferases were not measured at least 24 h postingestion were excluded. Hepatotoxicity was defined as peak serum aminotransferases >1,000 IU/L. Results: Sixty-two patients met our inclusion criteria. The median age was 21 years, and 41 patients (66%) were female. Fifteen (24%) co-ingested ethanol, and 7 (11%) were described as chronic alcoholics. Median time to treatment with IV NAC was 11.4 h [Interquartile range (IQR) 7.5–18.1] and median duration of treatment was 11.0 h (IQR 3.3–15.0). The median peak INR was 1.1 (IQR 1.0–1.2). Hepatotoxicity occurred in two patients (3.2%; 95% confidence interval 1.2–14.0). Anaphylactoid reactions were reported in 18 (29%) patients, of which 83% were cutaneous (facial flushing, urticaria, or pruritus). There were no deaths. Conclusions: In this cohort of patients receiving less than 20 h of IV NAC for acute APAP overdose, hepatotoxicity was infrequent. Anaphylactoid reactions were common and may have been a reason for discontinuation of IV NAC in some patients. Further research is required to identify patients at low-risk of hepatotoxicity who may benefit from a shortened course of IV NAC.

Accidental Unsupervised Ingestions (AUI) Most Common Type of Exposures Detected During Surveillance of Pediatric Exposures to Cough and Cold Medications 

Green JL, Reynolds K, Steitz AG, Dart RC, Pediatric Cough/Cold Medication Safety Surveillance Team.

Rocky Mountain Poison & Drug Center, Denver Health, Denver, CO, USA

Background: Safety concerns surrounding the use of cough/cold medications in children prompted the development of an ongoing surveillance system to monitor adverse events (AE) associated with these drugs. Methods: Using established methods, pediatric (age < 12 years) cases with AEs associated with the use of cough/ cold medications were collected from English language medical literature, National Poison Data System (NPDS), manufacturer safety records, FDA AERS and media reports from 2008 to 2009. The Pediatric Cough/ Cold Medication Safety Surveillance Team determined causal relationship between each reported drug and AE using predetermined definitions and then judged exposure dose, intent of administration and potential contributing factors for each case. Results: A total of 1,208 cases were reviewed by the Team, of which 916 were judged to have AEs at least potentially related to a cough/cold ingredient. Of the related cases, 552 (60%) were accidental unsupervised ingestions (AUI), 24% related to administration with therapeutic intent, 6% non-therapeutic intent and 10% unknown intent. Supratherapeutic dose was estimated in 88% of cases. Dose was unable to be determined in the remaining 12%. Age distribution of AUIs: 16% <2 years, 60% 2 to <4 years, 17% 4 to <6 years, 7% 6 to <12 years. Diphenhydramine was involved in 355 reports, including 3 deaths. These accounted for 64% of all AUIs (13% involved more than one product). Single substance exposures to singleagent diphenhydramine products accounted for 46% of all AUIs. AUI following therapeutic use of the product by a caregiver was reported in 5% of cases. Conclusions: AUIs account for the majority (60%) of all AE cases associated with cough/cold medication pediatric exposures detected in our system, perhaps due to the nature of the largest data source NPDS. These exposures primarily involve diphenhydramine products, likely because these products are widely used and available in homes throughout the US. Targeted interventions, such as proper storage of medication immediately after therapeutic use, are needed to reduce these preventable exposures that result in AEs in children.

Illicit Drug Exposures in Children 

Yin S, Bodmer M, Kokko J, Green JL, Dart RC.

Rocky Mountain Poison & Drug Center, Denver Health, Denver, CO, USA

Background: The epidemiology of illicit drug exposures in children is largely unknown. We sought to describe illicit drug exposures in children reported to US poison centers. Methods: The National Poison Data System was queried for all cases of illicit drug exposures in children <13 years of age from January 1, 2000 to July 31, 2009. Illicit drugs included the following American Association of Poison Control Center major categories: amyl nitrate, cocaine, gamma-hydroxybutyric acid (GHB), heroin, ketamine, methamphetamine, marijuana, hallucinogenic amphetamines, lysergic acid diethylamide (LSD), phenylcyclohexylpiperidine (PCP), other street drugs, unknown hallucinogens, other hallucinogens, unknown stimulants or street drugs. Cases were excluded if the exposure site was designated as a health care facility. Logistic regression was used to determine risk factors for a poor outcome (defined as death or major). Risk factors were exposure to each substance, exposure to >1 substance, intent, and age, which were adjusted against each other. Results: We identified 5,345 cases reporting 5,483 illicit drug mentions.

Factor Adjusted odds ratio 95% confidence interval
GHB 4.6 2.6, 8.0
Hallucinogenic amphetamines 3.4 2.1, 5.4
Cocaine 3.1 2.2, 4.3
PCP 3.1 1.8, 5.3
Age < 6 2.0 1.3, 3.0
Exposure to >1 substance 1.7 1.2, 2.4
Marijuana 0.5 0.3, 0.8

The annual average was 552 cases (range 507–629). Median age was 2 years (IQR 1.1, 5). Exposures primarily occurred at a residence (93%) and were coded as unintentional (78%). Marijuana (32%), methamphetamine (23%) and cocaine (21%) were the most commonly reported exposures: all other categories had less than 6.5% of exposures. Four percent of exposures resulted in death (n = 7) or major (n = 222) outcomes. An additional 805 cases were moderate exposures. The table shows factors significantly associated with death or major outcomes. Discussion: Abundant evidence exists to suggest children living in homes where drug abuse occurs are at risk for future drug use and psychological disorders. Since these exposures were largely unintentional and occurred in residences, this provides further evidence that an environment conducive to accidental illicit drug exposure is markedly unsafe for children and interventions reducing exposures to illicit drugs are needed.

Administration Routes Involved in Non-Medical Use of Long-Acting Opioids in the RADARS(R) System College Survey and Poison Center Programs

Buchholtz C, Thomasset C, Bailey JE, Dart RC, RADARS® System Poison Center Group.

Rocky Mountain Poison & Drug Center, Denver Health, Denver, CO, USA

Background: Among young-adult, non-medical opioid users, experimentation with alternate administration routes is frequently reported and used as a proxy for assessing escalation toward increasing use and drug dependence. The objective was to classify administration routes among college-aged non-medical users of longacting opioids (LAO) across two data collection systems. Methods: College students completed an online questionnaire (December 2009) for the RADARS System College Survey Program (CS), and were sampled equally from four US regions. Respondents answered questions about non-medical use and administration routes. The RADARS System Poison Center Program (PC) collects quality reviewed data weekly on acute drug intentional exposures from 48 of 60 US Poison Centers. CS and PC LAO cases (age 18–23) were identified. Of 1,936 CS cases, 2% (n = 41) involved at least one LAO. Acute LAO cases in CS were defined as those reporting past-month use less than or equal to 4 days (n = 31) and were compared to PC LAO intentional exposures (n = 264) as PC cases are acute in nature. Results: Forty-two percent of CS LAO cases reported two routes, and 34% reported three or more (n = 14). CS respondents can report multiple routes; swallowing whole (66% of cases), chewing (71%), inhalation (61%) and injection (24%). In PC LAO acute intentional exposures, only 0.8% involved two or more routes; the majority of cases involved swallowing whole (56%), with chewing, inhalation and injection comprising 16%. An independent samples Mann–Whitney U test revealed a significant difference between CS and PC LAO cases for the number of administration routes involved in acute cases (p < 0.001). Conclusions: Larger percentages of alternate administration routes (chewing, inhaling and injecting) reported for CS LAO cases suggests that CS may better capture experimental aberrant non-medical LAO use behaviors in this age group. A significant difference between the number of administration routes reported in CS and PC suggest these programs capture opioid use behaviors differently. An examination of both datasets provides better understanding of these behaviors than any one dataset alone.

Non-Fatal Attempted Suicide with Orpiment (Arsenic Trisulfide) Ingestion 

Tebb Z,1 Eberhard A,1 Heard K,2 Wendlandt R,3 Kosnett MM,2 Buchanan JA.2

1Department of Emergency Medicine, Rocky Mountain Poison & Drug Center, Denver, CO, USA; 2Rocky Mountain Poison & Drug Center, Denver Health, Denver, CO, USA; 3Department of Geology & Geological Engineering, Golden, CO, USA

Background: Arsenic trisulfide (As2S3), orpiment, is a mineral found near/in volcanic structures, coal combustion and herbal remedies. Minimal literature exists on orpiment ingestion. Case report: A 57-year-old male jeweler presented to an ED 13 h after he crushed and ingested an egg-size rock of orpiment in a self-harm attempt. The patient also took four acetaminophen/ diphenhydramine tablets and three melatonin pills. His only symptom prior to arrival was nausea. The patient was asymptomatic except for minor bleeding from a self-inflicted neck wound. Initial vital signs were HR 126 bpm, BP 162/105 mmHg, RR 18/min, RA Sat 98% and oral temperature 36.5°C. Physical examination revealed a clinically sober male with a 1.5 cm slash wound to his neck and was otherwise unremarkable. An ECG revealed a sinus tachycardia, HR 129 bpm with a QTc interval of 387 ms. A flat-plate X-ray showed hyperdense material throughout the bowel. Laboratory studies revealed normal LFTs, basic metabolic panel and an elevated white blood count of 20,700/μL. The patient was started on polyethylene glycol and observed in the telemetry unit. On Hospital Day 1, the patient remained asymptomatic and a random spot urine arsenic level was 1,489.5 μg/L (background range <30 μg/L). On HD 8, the urine arsenic level decreased to 800.2 μg/L. One month later he returned with a sample of the mineral and was asymptomatic. Discussion: We describe a case of As2S3 ingestion that resulted in minimal symptoms. The exposure was confirmed by an elevated urine arsenic concentration, an abdominal X-ray with radiopaque material and the product ingested was subsequently identified by a mineralogist with diffractometer and CuKa radiation. The LD50 of in rats and mice is approximately 10 times less than for arsenic trioxide (As2O3) and As2S3 is considered relatively non-toxic as it is comparatively insoluble with low bioavailability. As2S3 can oxidize to As2O3 on surface crystals which could lead to increased toxicity. Conclusion: We report a case of non-fatal As2S3 ingestion.

Characterization of Prescription Stimulant Exposures Using RADARS® System Poison Center Program Data 

Zosel A, Bodmer M, Buchholtz C, Bailey JE, Dart RC, RADARS® System Poison Center Group.

Rocky Mountain Poison & Drug Center, Denver Health, Denver, CO, USA

Background: Stimulant prescriptions for the treatment of attention-deficit/hyperactivity disorder have increased in recent years. Consequently, stimulant misuse and abuse is now a recognized problem in all ages. We describe stimulant exposures and associated medical outcomes as reported to the RADARS System Poison Center (PC) Program and compare to prescription opioid exposures. Methods: The PC Program captures weekly acute drug exposure data from 48 of 60 US PCs. These PCs cover 44 states (84% of US population). PCs use a standard electronic system to record calls from the public and the coordinating center performs quality control checks to verify coding accuracy. Stimulant (prescription amphetamine and methylphenidate) and opioid (buprenorphine, fentanyl, hydrocodone, hydromorphone, morphine, oxycodone, oxymorphone, tramadol) exposures from third quarter 2007 through third quarter 2009 were analyzed. Results: Thirty four thousand five hundred and forty (17 cases per 100,000 population) stimulant exposures (53% amphetamine; 47% methylphenidate) were reported over the study period. Mean age was 16.5 years (SD 13.5) and 57% were male. Site of ingestion was at own residence in most exposures (91%). The median number of substances ingested was one (range 2–26) while 31% (n = 10,379) involved two or more substances. Of known associated outcomes, 57% were no, minor or moderate effects, 2% (n = 524) were major effects and 0.06% (n = 21) were deaths. 119,475 opioid exposures (59 cases per 100,000 population) were reported over the study period; 5% (n = 5,878) were major effects and 0.5% (n = 572) were deaths. Thirty-two percent of stimulant exposures were intentional, compared with 57% of opioid exposures. Thirty-nine percent of stimulant exposures were therapeutic errors, compared with 22% of opioid exposures. Conclusion: Although fewer stimulant exposures were reported and were associated with fewer poor outcomes compared to opioid exposures, stimulant exposures still resulted in a significant number of poor outcomes. In addition, more therapeutic errors occurred with stimulants, reflecting the use of these drugs and associated dosing errors in young children. Our conclusions are limited to cases reported to PCs, which often under represent exposures.

 Spice Ain't So Nice

Banerji S, Deutsch CM, Bronstein AC.

Rocky Mountain Poison & Drug Center, Denver Health, Denver, CO, USA

Background: Spice, an incense product, is sold in the US and other countries. It may contain a variety of cannabinoid agonists like JWH 018 and JWH 073 that lack the classic cannabinoid structure and are therefore not controlled by the DEA or recognized as designer drugs. Recently, cases of Spice exposure via smoking have been under DEA, FDA, and CDC review. With the growing interest in the illicit use of Spice, poison centers have identified multiple cases of this emerging drug of abuse. We studied the incidence of Spice exposures in our five-state service region. Methods: Using NPDS, we queried for Spice cases using AAPCC Temporary Code #39 from September 2009 to April 2010. Audio review was performed on all cases except one.

Most frequent clinical effects
Symptom N (%)
Tachycardia 6 (66.7)
Anticholinergic toxidrome 4 (44.4)
Agitation/irritability 4 (44.4)
Tremor 4 (44.4)
Confusion 3 (33.3)
Pallor 2 (22.2)
Mydriasis 2 (22.2)
Hypertension 2 (22.2)

We used descriptive statistics to analyze the distribution of state, age, gender, clinical effects, and therapies. Results: Nine exposures and one information call were identified. Spice exposures were reported in three of our five states and 66.7% of exposures occurred in Colorado. All patients were male, median [min, max] age was 19 [13, 27] years. The most frequent clinical effect was tachycardia followed by the suggestion of anticholinergic toxidrome as described by the treating physician (Table). Seven (77.8%) patients received medical evaluation (six were in the hospital ED at the time of the call and one was referred by our center). Eight out of nine patients admitted to using Spice only. All symptoms resolved with symptomatic and supportive care including benzodiazepine administration (n = 3). Three patients had urine drug screens that were negative for THC. Conclusion: Spice appears to be an emerging public health problem among young males. The clinical picture is similar to THC exposure with some anticholinergic clinical effects. The incense’s availability and legal status make it easy for young people to acquire. This may burden health care facilities as most patients in our small sample required medical evaluation. Further characterization is needed with a larger sample size to better understand the toxicity of these THC homologs.

Which N-acetylcysteine Protocol is Associated with Better Outcomes?

Sivilotti MLA, Langmann C, Yarema MC, Juurlink DN, Johnson DW, Spyker DA, Thompson M, Green TJ, Dart RC, Rumack BH.

Queen’s University, Kingston, ON, Canada

Background: The three N-acetylcysteine (N-AC) protocols in widest use differ with regards to route (iv vs. po), dosing intensity (300 vs. 490 mg/kg during the first 24 h) and pattern (continuous vs. q4 h dosing). It remains unclear whether any of these differences affect efficacy. Methods: Using a large, structured medical records review of patients hospitalized for APAP poisoning in 34 Canadian hospitals from 1980 to 2005, we studied the association between initial N-AC protocol administered and peak INR, classified as ≤2, 2–4 and >4. Logistic regression was used to model coagulopathy with adjustment for age, alcoholism, and ethanol coingestion. We also corrected for overdose severity at baseline using the serum APAP multiplied by AST or ALT at the time of N-AC initiation, a predictor relatively independent of time or duration of ingestion. Subjects were classified based on the first NAC protocol initiated (20 h IV, 72 h PO, 48 h IV) even if the protocol was extended, abbreviated, or altered. Results: Of 11,987 hospitalizations, 4,075 were treated with N-AC and had sufficient laboratory data to model outcomes (peak INR >4 in 177, and 2–4 in 247 cases). There was a mild imbalance between groups in favour of the 72 h PO group, which had slightly younger patients and lower APAP × AT risk scores. After adjustment, the more dose intensive but intermittent protocols were associated with developing more severe coagulopathy, with the following cumulative odds ratios [95% CI]: 72 h PO 2.7 [1.4, 5.1]; 48 h IV 2.9 [1.5, 5.6] when compared with the 20 h IV group. The model fit was satisfactory (AUC 0.86), and the findings were similar using generalized logistic regression without assuming proportional odds. Conclusions: Initiating the 20 h IV protocol was associated with lower peak INR among all overdose types. While we cannot preclude confounding by severity, it appears unlikely that higher risk patients were systematically started on the higher dose protocols. Continuous rather than intermittent N-AC dosing may also account for the difference. These findings provide some reassurance to clinicians who continue to use the 150 mg/kg/day dosing of the terminal phase of the 20 h IV protocol, even for late presenting and chronic overdoses.

Therapeutic Errors Detected During Safety Surveillance of Over-the-Counter (OTC) Cough and Cold Medications in Children

Steitz AG, Green JL, Dart RC, Pediatric Cough/Cold Medication Safety Surveillance Team.

Rocky Mountain Poison & Drug Center, Denver Health, Denver, CO, USA

Background: Concerns have arisen regarding the safety of OTC cough/cold medication use in children. An ongoing multi-system surveillance model was developed to monitor adverse events (AE) associated with cough/cold drugs. Methods: Cases were identified from five sources: NPDS, FDA/AERS, English medical literature, news/media reports and manufacturer internal safety reports. Case inclusion criteria: age <12 years, exposure to ≥1 cough/cold ingredient, ≥1 AE which occurred in the US. The Pediatric Cough/Cold Medication Safety Surveillance Team met quarterly to determine causal relationship of each AE to each reported ingredient, to estimate dose and therapeutic intent, and to identify contributing factors. Analysis started in 3Q08 when all detection systems were active. Results: The Team reviewed 879 cases with event dates of 3Q08–4Q09. A total of 711 (81%) cases were determined as at least potentially related to a cough/cold ingredient: 707 non-fatal and 4 fatal (2 deaths in 3Q08, 1 in 4Q08 and 1 in 1Q09). 48% of children were age 2 to <4 years. While the majority of related cases were accidental unsupervised ingestions (66%), 146 (21%) of related cases involved therapeutic intent (drug administered to treat an illness). Of the therapeutic intent cases, 51% were estimated at supratherapeutic dose, 41% therapeutic dose and 8% unknown dose. Most adverse events were not serious and none resulted in death. Most common reported root causes for therapeutic intent exposure were wrong dose administered, exposure to combination products and exposure to multiple cough/cold products. Conclusions: Surveillance of AEs associated with cough/cold ingredient exposures in children indicates 1 in 10 AE reports involve therapeutic error. This suggests that interventions targeting prevention of unintentional overdose may be effective in reducing preventable dosing errors involving OTC cough/cold products.

Development of a Poison Center Based European Union Alerting System for Deliberate Chemical Release Detection

Schaper A,1 Desel H,1 Wyke S,2 Duarte-Davidson R,2 Bronstein AC.3

1GIZ-Nord Poisons Centre, Gottingen, Germany; 2Health Protection Agency, Oxford, UK; 3Rocky Mountain Poison & Drug Center, Denver Health, Denver, CO, USA

Background: Episodic health events may be due to isolated illnesses or part of a larger incident involving accidental or deliberate chemical release. Poison centers (PCs) receive information first hand involving both human and animal exposure data on suspected and confirmed poisonings. To date no standardized alerting system was available to communicate information on chemical health threats throughout the European Union (EU). In Europe, the European Commission (EC) has partially funded a consortium of six EU PCs (in six countries) to develop an Alerting System for the detection of chemical Health Threats (ASHT). The project developed from the Rapid Alert System for CHEMical health threats (RAS-CHEM). Methods: RAS-CHEM, a web-based system, has evolved over the last 30 months. RAS-CHEM is able to collect and analyze 19 data and free text fields, enabling EU PCs and other users to share information such as time, location, chemical substance, clinical effects (CEs) and event alert level. Results: RAS-CHEM was evaluated by entering 37 historical and simulated events from 14 countries: scenarios (n = 29) and historical mass intoxications (n = 8) that included the Bhopal release, the Halabdja Kurd attacks in Iraq and the Tokyo subway sarin attacks. The chemical classes included: gaseous (n = 10), liquid (n = 9), solid (n = 4), radioactive material (n = 9) and unknown (n = 5). Symptoms associated with these events included: respiratory (n = 13), neurological (n = 6), gastrointestinal (n = 4), local effects (n = 3) and miscellaneous (n = 11). The alert level was “for information” (n = 21), “urgent” (n = 6) and “very urgent” (n = 10). Conclusion: Testing demonstrated that PC case data can be entered into RAS-CHEM and shared with a variety of users across the EU to aid the detection, management and response to potential chemical public health threats. CEs can be readily categorized and events ranked by public health threat. RAS-CHEM has been specifically designed to be integrated into other EU Rapid Alert Systems [i.e. rapid alert system for dangerous consumer products (RAPEX)] and other international systems. The further development of RASCHEM has demonstrated the importance of PCs, and that they can play a key role as sentinels for the detection of chemical health threats.

Reduction over Time in RADARS® System Poison Center Opioid Abuse/Misuse Rates Associated with Prescription Monitoring Programs  

Reifler L,1 Droz D,2 Bucher-Bartelson B,1 Bailey JE,1 Schnoll S,3 Dart RC,1 RADARS(R) System Poison Center Group.1

1Rocky Mountain Poison & Drug Center, Denver Health, Denver, CO, USA; 2Ohio State Board of Pharmacy, Columbus, OH, USA; 3Pinney Associates, Bethesda, MD, USA

Background: Most states have implemented Prescription Monitoring Programs (PMPs) in an attempt to curb prescription drug abuse and diversion; however assessment of possible impact is only beginning. PMPs are statewide databases, containing patient-level prescription data on select drugs, intended for clinicians or other officials to use in identifying patients or providers engaging in abuse related illegal activities. Long acting opioids (which include extended release drugs) are preferred among prescription drug abusers, and PMPs’ impact on this type of abuse is of special interest. This analysis evaluated the association between PMPs and abuse and misuse rates over time for immediate release (IRO) versus long acting (LAO) formulations. Methods: Data (2003–2009) are from the RADARS® System Poison Center Program (PC) which collects quality reviewed intentional exposure (IE) events from participating US poison centers. PC IEs are considered surrogates of abuse and misuse. Formulations of oxycodone, fentanyl, hydrocodone, hydromorphone, morphine, and methadone were selected and summarized according to whether they were LAO or IRO. Information on states’ PMPs was compiled using public documents. Unique recipient of dispensed drug (URDD) data were used as a measure for drug availability in calculating IE rates. Repeated measures negative binomial regression was applied to predict states’ intentional exposure URDD rates. PMP presence was modeled as a time varying covariate for each state, and interactions of time, PMP status and LAO drug type were examined. Results: Both IRO and LAO results support that PMPs have an impact on IE rates over time. Model results display when states do not have a PMP in place, state IE rates increase on average 0.8% per quarter, where as rates decrease 1.2% (p = 0.0004) per quarter in those states with a PMP in place. However, results did not support that PMPs differentially influence LAO IE rates compared to IRO IE rates. Conclusions: PC observational data offer preliminary support that PMPs are effective, but do not support a difference in impact across long acting and immediate release opioid IE rates.

Health Care Facility Use of Poison Centers: 2000–2009

Spyker DA,1 Bronstein AC,2 Marcus SM.3

1Uniformed Services University of the Health Sciences, Bethesda, MD, USA; 2Rocky Mountain Poison and Drug Center, Denver Health, University of Colorado School of Medicine, Denver, CO, USA; 3New Jersey Poison Information & Education System, New Jersey Medical School, University of Medicine and Dentistry of New Jersey, Newark, NJ, USA

Introduction: The last decade has seen a dramatic increase in easy access to medical and toxicological information on the internet. We hypothesized health care facility (HCF) users of poison centers (PCs) might be using alternate information sources for managing poisonings. Methods: We examined HCF Human Exposure Calls (HCF-HECs) and HCF Information Calls (HCFICs) by day for the last decade (2000 through 2009) for secular trends (over Time) after accounting for Day of the week to account for weekly patterns (Day), Month of the year to account for seasonal patterns (Month), and 20 US holidays (Holiday). We compared these results). Increase and doubling time (DT = ln-2/ln-slope) and 95% confidence interval (CI) were calculated from logarithmic (proportional) models for the same parameters (Day, Month, Holiday and Time) using SAS JMP 6.0.0. Results: Day, Month, Holiday, as well as Time exhibited highly statistically significant (HSS, p < 0.0001) relations to both for HECs and ICs. The same was true for HFCHECs. HFC-ICs, however, did not show a steady increase over the last decade, rather a clear “inverted U shape” (second order) relation with Time with the peak occurring during 2005. These relations were likewise HSS. Conclusion: After accounting for the variation from Day, Month and Holiday, HCF-HECs (14.8% of HECs) are increasing more than twice as fast as all HECs (3.52 vs. 1.57% per year). This does not support our hypothesis that HCF staff are seeking alternative sources for poison management information. In contrast, HCF-ICs (2.88% of all ICs) are declining since about 2005. Thus, although HCF folks are calling more frequently for exposures, they may well be seeking alternative sources of information for the less frequent non-exposure queries. These results illustrate the use of a multivariate statistical model of the NPDS call data to answer a specific question. This approach may have application to PC surveillance, staffing, and funding.

Type of call Total calls (calls/decade) Average (calls/day) Percent of all calls Increase [95% CI] (%/year) Doubling time [95% CI] (years)
HCF exposure calls 3,532,458 967 14.8% 3.52 [3.59, 3.45]% 19.7 [19.3, 20.1]
HCF info calls 363,085 99.4 2.88%
All exposure calls 23,805,706 6,517 1.57 [1.52, 1.62]% 44.2 [42.8, 45.7]
All info calls 12,601,737 3,450 9.74 [9.65, 9.84]% 7.11 [7.05, 7.18]
Poison Center Call Time Metrics

Seroka AM,1 Lewis G,1 Banerji S,1 Green E,1 Spyker DA,2 Bronstein AC,1 Dart RC.1

1Rocky Mountain Poison & Drug Center, Denver Health, Denver, CO, USA; 3Uniformed Services University of the Health Sciences, Bethesda, MD, USA

Background: Poison Center (PC) calls have been shown to save health care dollars. There is little published data on the time required to manage a call. We studied call time for four call types: Exposure (E), Health Care Facility Exposure (HCFE), Information (I) and Drug Identification (DID). Methods: Talk Time and After Call Time were measured for 654 calls in April 2009 by eight non-clinical staff using stopwatches (Total Time = Talk Time + After Call Time). Six hundred and fifteen met quality criteria and formed the convenience sample. Timing was done by live observation or recordings (LR). Calls were taken by Agents: specialists in poison information (SPIs) and poison information providers (PIPs). The process was validated by comparing Time of Day (ToD) and Talk Time to call time obtained from our Avaya Call Management System v14.0. Call time measures (Type, Agent, Timer, ToD, Date, L-R) were examined using bivariate and multivariate least squares methods applied to linear or logarithmic (proportional) models as appropriate. For log models, geometric means [95% CI] provided point estimates and statistical analyses used SAS JMP v6.0.0. Results: Of the 615 cases, 21 SPIs handled 470 (76%) and 7 PIPs handled 145 (24%). Most, 557 (91%) were from live timings and 58 (9%) from recordings. Both Talk Time and Total Time were log-normally distributed and statistical analyses were carried out using the log models. Final models for both Talk and Total Times included only Call Type and Agent. Distinguishing between SPI-PIP did not improve the model fit, but differences among the 28 Agents were statistically significant (p < 0.0001). Conclusion: As expected, HCFE calls took the most time, followed by E, I, and DID. A limitation was that our SPIs manage multiple calls simultaneously and cannot always chart after each call. Call timing data is valuable to PC managers to build staffing models, evaluate performance and determine service cost and pricing.

    Arithmetic median (min, max) Geometric least squares mean [95% CI]
Type of call Number Talk time (min) Total time (min) Talk time (min) Total time (min)
HCFE 99 7.22 (0.88, 25.7) 9.47 (1.88, 29.8) 7.51 [6.43, 8.78] 9.59 [8.3, 11.1]
E 363 3.6 (0.43, 23.3) 5.42 (1.03, 33.5) 4.01 [3.52, 4.55] 5.87 [5.21, 6.6]
I 75 2.48 (0.53, 15.9) 4.18 (0.93, 22.4) 2.87 [2.37, 3.47] 4.02 [3.38, 4.79]
DID 78 1.16 (0.42, 6.22) 1.38 (0.42, 7.2) 1.46 [1.2, 1.79] 1.61 [1.34, 1.94]
Total 615        
Antidote Stocking in Denver Metro Hospitals

Tate I,1 Buchanan JA,2 Bogdan GM,2 Chuang R.2

Department of Emergency Medicine, Denver Health, Denver, CO, USA; 2Rocky Mountain Poison & Drug Center, Denver Health, Denver, CO, USA

Background: Poisonings and drug overdoses can be successfully treated when antidotes are administered in a timely fashion. The Antidote Summit Authorship Group published updated guidelines for stocking of antidotes in 2009. The objective was to examine whether hospitals in metropolitan Denver possess adequate antidote stocks according to these new guidelines. Methods: Fourteen antidotes were selected for evaluation from 24 recommended for stocking: 10 antidotes were excluded from our survey because they were multi-purpose medications and likely universally stocked. For crotalid envenomation and cyanide toxicity, 2 antidotes were listed but stocking only one was recommended. Surveys were sent to pharmacists at 28 local hospitals between January and February 2010 asking for their current stocks of the 14 selected antidotes. Results: Eleven of 28 (39%) hospitals responded to the survey. A fully stocked pharmacy would have adequate supplies of 12 antidotes since 2 were redundant. None had adequate stocks of 12 recommended antidotes. The 11 respondents had adequate stocks for an 8-h treatment course for a mean of 7 of 12 antidotes (58%) and for a 24-h treatment course for a mean of 6 of 12 antidotes (50%). The most commonly stocked antidote was the cyanide antidote kit (amyl nitrite, sodium nitrite, sodium thiosulfate) and all hospitals had adequate supplies for 24-h courses. N-acetylcysteine was the second most stocked antidote, with all hospitals possessing supplies for an 8-h course and 10 of 11 (91%) possessing supplies for a 24-h course. The least commonly stocked antidotes were crotalidae antivenin (Wyeth Pharmaceuticals, no longer in production) and calcium trisodium pentetate (calcium DTPA), both of which were not stocked by any hospitals. Discussion: The sample of hospitals responding to this survey represented a broad spectrum of institutions and thus their stocking practices are likely representative of hospitals in the area. Though local hospitals did not meet recommended guidelines for antidote stocking, commonly used antidotes such as Nacetylcysteine were available. Conclusion: We suggest pharmacy directors, emergency physicians and hospital administrators review the Antidote Summit Authorship Group recommendations and update antidote stocks accordingly and considering local needs and hazards.

Assessment of Toxicology Knowledge in Fourth Year Medical Students 

Windels D,1 Heard K,2 Druck J,1 Buchanan JA.2

1University of Colorado Denver School of Medicine, Aurora, CO, USA; 2Rocky Mountain Poison & Drug Center, Denver Health, Denver, CO, USA

Background: Pharmacology and toxicology are core content knowledge for physicians. Medical students should demonstrate understanding of general pharmacology and basic treatment of poisoning. The objective of this study was to measure the knowledge of fourth year medical students (MS4) on these topics. Methods: A multiplechoice survey (15 questions) was administered to MS4. Questions were grouped into three categories: treatment of poisoning, pharmacokinetics and pharmacologic effects. Students were grouped in by intended specialties: pharmacologic intense (anesthesia, emergency medicine, internal medicine, pediatrics and psychiatry); less pharmacologic intense (dermatology, OB/GYN, ophthalmology, pathology, physical medicine/rehabilitation, radiology and surgery); or no specialty recorded. Students were also grouped by completion of a clinical pharmacology and/or toxicology elective or neither elective. Groups were compared using ANOVA. Results: A total of 108 of 136 students completed the survey. Students completing the toxicology elective had higher mean scores than those taking neither elective; however, the scores for pharmacology intense specialties were not different from less pharmacology intense specialties. Discussion: Performance on this test appears to be improved by completing a toxicology vtelective. Performance was not clearly higher for students planning on a pharmacologic intense specialty. A limitation is that the study sampled a small population and covered minimal material. Conclusion: Data from this study suggests MS4 are lacking in core content knowledge related to toxicology. Implementation of required courses focused on toxicology may improve performance.

And the Survey Said! Use of Web-Based Survey as Education Tool in a Poison Center

Banerji S, Bronstein AC.

Rocky Mountain Poison & Drug Center, Denver Health, Denver, CO, USA

Background: During routine case review, we identified several cases with extended antidote therapy. Although patients had good outcomes, the reasons for the therapy extensions were unclear. When the Specialists in Poison Information (SPI) were asked about their treatment decisions, a variety of answers was returned. Recognizing an educational opportunity, we decided we needed a simple way to characterize our baseline (current) state. Instead of a formal lecture, we opted to poll the SPIs on their current management understanding using a standard internet survey tool. Methods: We selected as our survey platform based on ease of use, functionality, and expense (no cost associated with basic membership). A hypothetical teaching case involving a situation where the antidote was extended inappropriately was created with seven survey questions. SPIs were guided through the case and asked to select the action which most closely resembled their own case management practices. Specialists were given 1 month to complete the survey. Time off the phone to complete the survey was not given to the SPIs. The survey link was emailed to staff and a reminder email was sent 2 weeks before the deadline. Individual results were kept anonymous as they were not required to “sign in.” A post-survey analysis was presented to the SPIs. Results: A total of 25/33 (76%) SPIs completed the survey by the deadline and 1 SPI after the deadline. Half of SPIs agreed with the original course and management of the case and the other half answered they would have managed the case differently. 17/26 (65%) of SPIs wrote comments which helped explain their rationale for the decisions they made. Estimated time to complete survey was 5–10 min. We presented a formal lecture on the findings and reviewed management rationale. Changes to our current management guidelines were made as a result of the survey findings. Conclusion: Web-based tools are frequently utilized in many work environments. The use of a web-based survey in a poison center can provide a valuable platform for educational and practicebased applications. Our survey results showed that re-education was needed on some key management issues. Surveys can be non-threatening, cost-effective and fun for participants.

Categories/subcategories Number of calls AL (%) Number of calls national (%) p Value
Unintentional exposures 60.83 73.54 p < 0.0001
Unintentional misuse 5.07 10.35 p = 0.02
Intentional exposures 31.80 19.83 p < 0.0001
Intentional suspected suicide 25.80 13.85 p < 0.0001
A Decade of National Poison Data System (NPDS) Call Data – Baseline Statistical Models 

Spyker DA,1 Bronstein AC.2

1Uniformed Services University of the Health Sciences, Bethesda, MD, USA; 2Rocky Mountain Poison and Drug Center, Denver Health, University of Colorado School of Medicine, Denver, CO, USA

Introduction: NPDS call volume data exhibits repeated patterns by year, by week and, changes with some holidays. Understanding these patterns is critical to interpretating NPDS call volume over time. We developed a baseline statistical model to support the quantitative analyses of NPDS call data. Methods: We examined all Human Exposure Calls (HECs, N = 23,805,706) and all Information Calls (ICs, N = 12,601,737) by day for the last decade (2000–2009) for both linear and log relationship to: Day of the week to account for weekly patterns (Day), Month of the year to account for seasonal patterns (Month), 20 US holidays (Holiday), as well as Time to account for overall secular trends. Day, Month, and Holiday were each treated as nominal (not ordinal or continuous) variables to avoid assumptions about their relative contributions. via SAS JMP v 6.0.0. Results: Linear were better than log models for both HES and IC so subsequent results refer to the linear models, though doubling times were based on log models. Day, Month, Holiday and Time exhibited highly statistically significant (HSS, p < 0.0001) relations to both HES and IC. Most of the individual Days, Months and Holidays (elements) were statistically significant (SS, p < 0.05). All but one of the elements which were not SS in the HES model were SS for IC, and vice versa. Thus all elements were included in both models. The slope [95% confidence interval] and associated doubling time (DT) were: EC slope = 98.7 [95.5, 102] calls/day/year, DT = 44.2 [42.8, 45.7] years IC slope = 314 [312, 317] calls/day/year, DT = 7.11 [7.05, 7.18] years Conclusions: Even after accounting for the variation from Day, Month and Holiday, the secular trend (Time) remained HSS for both HESs and ICs, with a greater rate of increase for the ICs. We have not included singular events (disasters or public health issues). Our findings may have application to PC surveillance, staffing, and funding. These initial results underscore the importance of considering all “statistically important” contributors in any quantitative analyses of NPDS call data.

Contact Us

For any business inquiries for the Drug Center or the Poison Center, please email us at or call  (866) 871-4980.

In case of poisoning or chemical exposure, dial (800) 222-1222 and speak directly to a poison control specialist. If you are experiencing an emergency, please dial 911.

For any media inquiries, please email