Hrev_master [page 14] [Emergency Care Journal 2017; 13:6302] A retrospective analysis of the impact of toxicological diagnostics on clinical decision making in cases of acute drug poisoning Janne H. Liisanantti,1 Suvi Lehtiniemi,2 Tero I. Ala-Kokko2 1Division of Intensive Care Medicine, Department of Anesthesiology, Oulu University Hospital; 2Anesthesiology and Intensive Care, Oulu University Hospital, Oulu, Finland Abstract The outcome of acute drug poisoning is good. In only a few occasions specific treat- ments are needed. Toxicological screenings are recommended when acute drug poison- ing is suspected. In this retrospective obser- vational study the impact of routine screen- ing on treatment decisions was analyzed. All patients with acute drug poisoning admitted to the emergency department of our university hospital during one year (2013) were retrospectively analyzed. The patients were categorized into two groups: those who received specific therapies due to the poisoning and those who received only symptomatic treatment. Results: there were a total of 318 cases with acute drug poison- ing of which 120 led to intensive care treat- ment. Toxicological screening was per- formed in 225 cases (70.8%). The screening tests were more often taken from patients who were unconscious (89%) or had altered consciousness (79%) than from patients with normal consciousness (63%, P<0.001). The rate of specific treatment was higher among screened patients compared with patients without screening (18.7 vs 1.1%, P<0.001). However, unexpected screening results were found in 37 of the 225 screened patients out of whom only 6 cases received specific treatment. Most patients with acute drug poisoning were toxicologically screened, but every sixth had an unexpected finding. The rate of patients with unexpect- ed screening result receiving specific treat- ment was low. Introduction Acute drug poisoning is a relatively common cause of emergency department (ED) admissions. The outcome in hospital care is good even with the cases treated in the intensive care unit (ICU) with the mor- tality rate less than 3%,1-4 and in a Norwegian study by Lund et al the hospital mortality was even less, 0.8%.5 The hospital length of stay (LOS) is usually less than one day.1-5 The causes of intoxication include analgesics and psychotropic drugs, such as antidepressants and anxiolytics.1,5 In Finland the most typical cause of acute drug poisoning is the intake of psychotropic products in association with alcohol intoxi- cation.1 Similar findings have been reported from Oslo, Norway, but for instance in a Scottish series paracetamol was the most frequent cause of poisoning.5,6 The recommended cornerstones in the management of patients with acute drug poisoning are stabilization of the vital func- tions and the use of activated charcoal to prevent the absorption of the ingested agent. Toxicological screenings can be used for diagnostics and to guide the management of the patients.7,8 The diagnostics in acute poi- soning is challenging as shown in one Finnish and one Norwegian study where the suspected causes of poisoning and the results of the toxicological screenings did not correlate well with patient history.9,10 However, the evidence supporting the use of toxicological screenings is lacking. There are studies questioning the use of screen- ings, but they are with limited number of patients or including also not poisoned patients11-13 or conducted with pediatric population.14 The outcome of the patient with acute drug poisoning is good, but the impact of the used screenings on the outcome is unclear. Therefore, we decided to retrospec- tively analyze the impact of the performed toxicological screenings on the decisions to start specific therapies or extracorporeal treatment. Materials and Methods Setting and study design This retrospective observational study was conducted in Oulu University Hospital located in Northern Finland. Due to the ret- rospective study design no statement was obtained from the hospital’s ethics commit- tee. The protocol was approved by the North-Ostrobothnian hospital administra- tion. All the patients with acute drug poison- ing admitted to the hospital ED between January 1, 2013, and December 31, 2013, were included in the study. According to the local guidelines, poisoned patient is admit- ted to ICU if there is a need of airway man- agement, if there is a risk of the need of air- way management, need of vasoactive drugs or need of specific extracorporeal treatment for acute poisoning. In other cases patients are admitted to normal wards. The patients were identified from the hospital’s dis- charge registry using ICD-10 diagnosis codes T36-T50.9. The patient data were manually retrieved from the medical records and hospital’s laboratory system. The data consisted of patient demographics, including hospital LOS; complications, such as pneumonia, seizures, rhabdomyoly- sis and liver failure; and hospital mortality. The cause of poisoning was determined using the screening results, and if these were lacking, the most likely cause was recorded on the basis of the patient’s med- ical history. Patients were categorized as conscious when Glasgow Coma Scale (GCS) score was 14-15, as having altered level of consciousness when GCS score was 9-13, and as unconscious when GCS score was less than 9. Poisoning severity score (PSS) was determined using medical records. Poisoning was considered severe if PSS was above 2.15 The patients were divided into two groups by the need of specific therapies: i) patient received specific therapies, includ- ing N-acetyl cysteine (NAC) infusion, or forced alkaline diuresis; or the patient received extracorporeal treatment, includ- ing acute hemodialysis, acute hemodiafil- tration, hemoperfusion, or molecular adsor- Emergency Care Journal 2017; volume 13:6302 Correspondence: Janne H. Liisanantti, Division of Intensive Care Medicine, Department of Anesthesiology, Oulu University Hospital, 90029 Oulu, Finland. E-mail: janne.liisanantti@ppshp.fi Key words: Acute drug poisoning; Extracorporeal treatment; Diagnostics; Toxicological screening. Contributions: Janne Liisanantti, Suvi Lehtiniemi and Tero Ala-Kokko equally con- tributed in design of the study, data analysis, and writing of the manuscript. Suvi Lehtiniemi and Janne Liisanantti retrieved the data from the medical records Conflict of interest: the authors declare no potential conflict of interest. Received for publication: 22 September 2016. Revision received: 23 March 2017. Accepted for publication: 24 March 2017. This work is licensed under a Creative Commons Attribution 4.0 License (by-nc 4.0). ©Copyright J.H. Liisanantti et al., 2017 Licensee PAGEPress, Italy Emergency Care Journal 2017; 13:6302 doi:10.4081/ecj.2017.6302 Non co mmerc ial us e o nly bent recycling system (MARS); ii) patient received symptomatic therapy only, includ- ing management of vital functions. Use of activated charcoal was included in sympto- matic therapy. Toxicological diagnostics Hospital laboratory provides an immune chemical toxicological screening set for urine amphetamine, methampheta- mine, benzodiazepines, cannabinoids, cocaine, opioids, and buprenorphine. Paracetamol poisoning is screened using homogenous enzyme-immunoassay method that produces a measurement of blood paracetamol concentrations. Siemens Advia Chemisty XPT -device is used for the analy- ses (Siemens Healthcare GmbH, Erlangen, Germany). There are no standard guidelines for toxicological screening in the unit. Statistical analysis The statistical analysis was performed using IBM SPSS Statistics22 software. Continuous variables are expressed as medians with 25th and 75th percentiles unless otherwise stated, and proportional data are expressed using total number (n) and percentage (%). Differences were con- sidered to be statistically significant with P value less than 0.05. The statistical signifi- cance of proportional data was tested using Pearson chi-square, and medians between the groups were tested using nonparametric Mann-Whitney test. Results There were a total of 32,089 admissions that led to hospitalization in the ED during the study period, and in 318 of these admis- sions (1.0%) the patient had the discharge diagnosis of acute drug poisoning and was included in the study. Of these 318 patients, 169 (53.1%) were males, and the median age of the study population was 33.0 years (25.8-47.0). A total of 120 hospital admis- sions (37.7%) led to an ICU admission. The median hospital LOS was 1 day (1-2) (Table 1). The toxicological screening from urine was taken in 202 cases (63.5%) and blood paracetamol concentration was screened in 225 (70.8%) cases. Urine samples were also screened from all patients screened for paracetamol; thus, the total number of screened patients was 225, which was 70.8% of all patients admitted to the ED due to acute drug poisoning. The screening tests were more often taken from unconscious patients (46 of 52, 88.5%) or patients with altered consciousness (67 of 85, 78.8%) compared with patients who had a normal level of consciousness (114 of 180, 63.3% P<0.001). The screened patients were more often admitted to the ICU than the patients who were not screened (106 of 225, 47.1% vs 14 of 93, 15.1%, P<0.001) (Table 2). The specific therapies or specific extra- corporeal treatment was required in 42 (17.5%) of the screened 225 cases. The most common required specific therapy was NAC infusion, which was used in 31 admis- sions (9.7% of all the 318 admissions). Forced alkaline diuresis was used 9 times (2.8% of all 318 admissions). Hemodialysis was used 4 times (1.3% of all 318 admis- sions), hemodiafiltration was used twice, and one patient received MARS therapy. Patients requiring special therapies or spe- cific extracorporeal treatment had a higher rate of paracetamol ingestions than patients who received symptomatic therapy only, (31 of 42 [73.8%] vs 4 of 183 [2.2%], P<0.0001), and they also had longer median ICU stay (28 hours vs 12 hours, P<0.0001) and longer median hospital stay (1.5 days vs 1 day, P=0.001). Patients with symptomatic Article Table 1. Demographics of the 318 patients admitted to hospital due to acute drug poisoning. Total Patients without Patients with P N=318 toxicological screening (N=93) toxicological screening (N=225) Gender f/m 149/169 46/47 122/103 0.549 Age, median, y 33.0 [25.8-47.0] 36.0 [26-52] 32 [26-46] 0.166 Suicide or DSH 239 72 [79.1] 167 [75.2] 0.461 Level of consciousness at ED Unconscious; GCS <9 52 (16.3) 6 (6.5) 46 (20.5) <0.001 Altered level of consciousness; GCS 9-13 85 (26.8) 17(18.3) 68 (30.4) Conscious; GSC 14-15 180 (56.6) 70 (75.3) 110 (49.1) Cause of the poisoning Atypic neuroleptics 54 (17.0) 22 (23.7) 32 (14.2) 0.006 Benzodiazepines 43 (13.5) 12 (12.9) 31 (13.8) Paracetamole 35 (11.0) 0 35 (15.6) Antiepileptics 27 (8.5) 7 (7.5) 20 (8.9) SSRI 21 (6.6) 5 (5.4) 16 (7.1) Opioids 21 (6.6) 7 (35.5) 14 (6.2) Neuroleptics 15 (4.7) 5 (5.4) 10 (4.4) TAD 11 (3.5) 2 (2.2) 9 (4.0) Other 91 (28.6) 33 (35.5) 58 (25.8) Multiple ingestion 252 (79.2) 66 (71.0) 186 (82.7) 0.270 Specific therapy or extracorporeal treatment 43 (13.5) 1* (1.1) 42 (18.7) Intubation 45 (14.2) 5 (5.4) 40 (17.8) 0.064 PSS>2 20 (6.3) 5 (5.4) 15 (16.7) 0.660 Hospital LOS, median, d 1 (1-2] 1 (1-2) 1 (1-2) 0.270 ICU admission 120 (37.7) 14 (15.1) 106 (47.1) <0.001 ICU LOS, median, h 14 (8-20) 15 (6-17) 13 (9-20) 0.889 Complication 25 (7.9) 4 (4.3) 21 (9.3) 0.129 ED, emergency department; GCS, Glascow Coma Scale; ICU, intensive care unit; LOS, length of stay; PSS, poisoning severity score; TAD, tricyclic antidepressants *One patient received NAC for indication not parac- etamol poisoning [Emergency Care Journal 2017; 13:6302] [page 15] Non co mmerc ial us e o nly [page 16] [Emergency Care Journal 2017; 13:6302] therapy were more often unconscious at ED [101 of 183 (55.5%) vs 13 of 42 (31.0%), P=0.004]. There were no differences in unexpected screening results, the rate of complications, or admissions to ICU between the groups. Unexpected findings were recorded in 37 (16.4%) of the screened 225 patients, and 6 (2.7%) of them were among those 42 patients who required specific therapies or specific extracorporeal treatment. Three of these were paracetamol poisonings recog- nized by the screening. The other three were identified using specific laboratory tests due to clinical suspicion (Table 3). In 178 cases (78.0%) the results of the performed screenings or diagnostic tests were commented in the medical records made in the ED. All patients had a good hospital out- come. Complications were recorded in 25 patients (7.8%), including 9 cases of seizures (2.8%), 8 cases of pneumonia (2.5%), 5 patients who had a prolonged stay in the hospital due to altered consciousness, 2 patients who developed a liver failure, and 1 patient with rhabdomyolysis. The record- ed complications were not related to the toxicological screenings used; there were 4 (4.4%) recorded complications among non- screened patients and 21 (9.2%) recorded complications among screened patients (P=0.148) Discussion The main finding of the present study was that 70% of the admissions due to poi- soning involved screening but only one- fifth of the screened patients received spe- cific care or extracorporeal treatment. Moreover, the rate of unexpected screening results leading to specific care or extracor- poreal treatment was less than 3% (6 of 225). We evaluated a non-selected patient population in a university hospital setting aiming to study the usefulness of toxicolog- ical screenings to detect the need of specific care or extracorporeal treatments. The pre- vious studies focusing the use of screenings have questioned the impact of toxicological screenings in poisoned patients.11-14 However, in these papers the patients’ char- acteristics differenced from the present study including lower number of uncon- scious patients and other than poisoned patients.11-13 In the present study, the patient demo- graphics, causes of admissions, rate of ICU admissions, and the outcome are compara- ble with the Norwegian study by Lund et al published in 2012.5 Corresponding with this Norwegian study, the need for specific ther- apies or extracorporeal treatment was low also in our material. We did not find differ- ences in outcome between the patients who required specific therapies or extracorporeal treatments and those who needed sympto- matic therapy, but the median hospital LOS and ICU LOS among patients with sympto- matic care were shorter. Also, the rate of unexpected screening results was equal Article Table 2. The 225 toxicologically screened (toxicological urine samples or paracetamol screening) drug poisoned patients divided into the two groups; need of specific therapies of extracorporeal treatment (N-acetyl cysteine or forced alkaline diuresis, acute hemodialysis, hemodiafiltration, or molecular adsorbent recirculating system) or symptomatic care. Specific therapies or extracorporeal treatment Symptomatic therapy P N=42 (17.5) N=183 (81.3) Age, median, y 30 [21-52] 32 [26-46] 0.774 Gender f/m 22/18 77/104 0.101 Unexpected finding in the screening 6 (14.3) 31 (16.9) 0.676 or in the specific laboratory test Unconscious at ED 13 (31.0) 101 (55.5) 0.004 Suicide and DSH 32 (78.0) 135 (74.6) 0.643 Specific therapies NAC 31 (76.9) Hemodialysis 4 (10.0) Hemodiafiltration 2 (5.0) Forced alcaline diuresis 9 (22.5) MARS 1 (2.5) Cause of the poisoning Paracetamol 31 (73.8) 4 (2.2) <0.0001 Atypic neuroleptics 0 32 (17.5) Benzodiazepines 0 31 (16.9) Antiepileptics 1 (2.5) 19 (10.4) SSRI 0 16 (8.7) Opioids 0 14 (7.7) Neuroleptics 1 (2.4) 9 (4.9) TAD 2 (4.8) 7 (3.8) Other 7 (16.7) 51 (27.9) Multiple ingestion 32 (76.2) 154 (84.2) 0.219 Intubation 4 (9.1) 41 (14.9) 0.242 PSS>2 4 (9.5) 16 (5.8) 0.358 ICU admission 20 (47.6) 86 (47.0) >0.9 ICU LOS, median, h 28 [15.5–45.5] 12 [8-18] <0.0001 Hospital LOS, median, d 1.5 [1-3] 1 [1-2] 0.001 Complication 6 (14.3) 15 (8.2) 0.221 DSH, deliberate self-harm; ED, emergency department; ICU, intensive care unit; LOS, length of stay; MARS, molecular adsorbent recirculating system; NAC, N-acetyl cysteine; PSS, poisoning severity score; SSRI, selective serotonin reuptake inhibitors; TAD, tricyclic antidepressants; PSS, poisoning severity score. Non co mmerc ial us e o nly [Emergency Care Journal 2017; 13:6302] [page 17] between patients who required specific therapies or extracorporeal treatment and those with symptomatic care despite the dif- ferences in agents ingested between the groups. This result is interesting when com- pared to the results of Pohjola-Sintonen at al and Heyerdahl et al, which showed that there were high discrepancies between clin- ical and laboratory assessments most often with products affecting central nervous sys- tem.9,10 These discrepancies were consid- ered clinically important in 11% to 18% of the cases. In our series 16% of the screening results were considered unexpected. However, in our study the rate of unexpect- ed screening results leading to specific treatment was less than three percent. Also in the previous Finnish series by Pohjola- Sintonen et al the discrepancies that were considered clinically significant did not lead to specific therapies and were not relat- ed to recorded complications.9 In the pres- ent results it is notable that only 3 of the 6 unexpected findings among patients who required specific therapies were detected by the routine screening (paracetamol) and in only 2 of these cases the patient had a low- ered level of consciousness. The detection of the other 3 cases required specific labora- tory testing (lithium and salicylate concen- tration). In the light of the present results the use of toxicological screening without suspicion of poisoning requiring specific therapies or extracorporeal treatment can be questioned. In most cases of the ICU admit- ted patients the screening did not lead to in any specific treatment. The decision to admit the patient to ICU was more related to vital functions than the screening results. Furthermore, the present study population included only 6% of severe poisonings according to PSS and thus the role of toxi- cological investigations could be question- able. The history of substance abuse of hos- pitalized patients can be unreliable. In a study conducted in Tennessee, United States, a total of 1500 ED patients were interviewed, and the interviews were com- pared to toxicological screening results. This comparison revealed discrepancies in self-reported and screened abuse of opioids, benzodiazepines, cannabis and stimulants. However, it is notable that the patients included in the study had not been admitted to hospital for poisoning-related reasons.16 According to the present results, instead of the diagnostics of the known or suspected drug-poisoned patients, the screenings could be used in differential diagnostics of altered level of consciousness when other severe causes such as neurological diseases are ruled out. This way the screening may have an impact on care but that can also be debated.12 In most cases the poisoning is a self-correcting condition, in which the body requires time to metabolize and eliminate the agent.5 In the present study, the screen- ings were more often used in patients with altered level of consciousness. This can lead to bias in the results since specific treat- ments also less often needed among these patients. It would have been interesting to see the impact of screenings on the patients management if all the patients would have been screened. However, the aim of the present study was to also to evaluate the use of the screenings, not only the impact of their use. The outcome in the present study was good and thus similar to the outcomes reported previously.2,5 There are recommen- dations that support the use of toxicological screenings for diagnostics in the patients admitted to hospital due to acute drug poi- soning or in cases of suspected acute drug poisoning.7,8 We were not able to show any major differences in the outcome or the rate of complications between the screened and non-screened patients. The screened patients more often presented altered level of consciousness and were admitted to the ICU, but on the other hand, two-thirds of the patients who required specific care were not unconscious at ED while the correspon- ding rate among those who needed sympto- matic care was less than half. Considering the very low mortality and morbidity among hospitalized patients with drug poi- soning, it may be asked how the outcomes of these patients may be further enhanced. The screenings would probably have had an impact on the care in every fifth screened patient (n=42) if the specific therapies or specific extracorporeal treatment had been conducted according to screening results. If only taking into account patients with unex- pected findings, only every seventieth screened patient (n=3) benefitted from the screening, which is a very small number of patients. With limited number of patients receiving specific care as well as limited number of patients with severe poisonings the collaboration with clinical toxicologists is important when producing guidelines concerning the use of screenings to imple- ment the right diagnostic tests. The benefit that is possible to gain using screenings seems to be minor in this patient group with good outcome, limited number of specific treatment procedures, and rela- tively simple, symptomatic care. Similarly to our study patients, most of the patients in the Norwegian cohort did not receive any other treatment than supportive therapy; a minority received activated charcoal and special procedures to enhance the elimina- tion of the harmful agent, and antidotes were used in also less than 50% of the patients.5 The focus should be in prevention of complications and factors associated with prolonged ICU stay. Our previous study showed that prolonged ICU stays Article Table 3. Patients (n=6) who received special therapies or extracorporeal treatment with unexpected screening or laboratory test results. Gender, Agent Consciousness Intake Laboratory ICU Hospital Specific Poisoning age ingested at ED motivation test LOS LOS therapy/ severity (hours) (days) extracorporeal score treatment f, 66 Lithium Conscious Accident Blood 120 6 Hemodialysis 2 lithium concentration f, 52 Paracetamol Altered level Suicide Paracetamol screening 28 4 NAC 2 of consciousness f, 69 Salicylate Conscious - Blood salicylate 45 7 Hemodialysis, forced 2 concentration alkaline diuresis f, 46 Lithium, multiple Unconscious Suicide Blood lithium 46 3 Forced alkaline diuresis 3 ingestion concentration f, 76 Paracetamol Conscious Suicide Paracetamol screening - 1 NAC 1 f, 20 Paracetamol Unconscious Recreational use Paracetamol screening 8 3 NAC 2 ED, emergency department; ICU, intensive care unit; LOS, length of stay; NAC, N-acetyl cysteine. Non co mmerc ial us e o nly [page 18] [Emergency Care Journal 2017; 13:6302] were associated with organ dysfunctions, such as hypotension and respiratory failure.2 The screenings may be useful in cases of potentially dangerous poisonings, such as lithium or paracetamol poisonings with specific treatments available, but according to our results, only with clinical suspicion. However, the present data as a single center, retrospective data should be considered as hypothesis generating. The findings call for a large prospective study to examine the impact of toxicological screen- ings on the treatment decisions, outcome and whether these tests could reduce the number of ICU admissions in poisoned patients without need of organ support, including airway management. We present- ed data of 318 patients with acute poisoning who were admitted to university hospital ED within one year. According to our knowledge this was the first study analyz- ing the impact of toxicological screening on the management of patients with acute drug poisoning. We included all drug-poisoned patients, mimicking the actual clinical situ- ation. The present results are comparable with the Norwegian study in terms of patients’ characteristics, including the level of consciousness on admission, need of ICU admission, and outcome, which promotes the generalizability of the present results at least to Nordic countries.5 There are limita- tions in this study. First, the design was ret- rospective. However, with this setting we were able to show afterwards the decisions that were made using the screenings. Also, the number of patients was limited, but we covered one year’s admissions to a univer- sity-level hospital. A study of the impact of screenings on mortality would have required inclusion of several hundred patients, which would have expanded the time span too much. The aim of the present study was not to study the impact of screen- ings on outcome but the impact of the screenings on the treatment decisions. We were not able to analyze the use of screen- ing in differential diagnostics in the patient population that showed an altered level of consciousness at the ED. Conclusions In conclusion, most patients with acute drug poisoning were toxicologically screened, but every sixth had an unexpected finding. The rate of patients with unexpect- ed screening result receiving specific treat- ment was low. References 1. Lapatto-Reiniluoto O, Kivisto KT, Pohjola-Sintonen S, et al. A prospective study of acute poisonings in Finnish hospital patients. Hum Exp Toxicol 1998;17:307-11. 2. Liisanantti JH, Ohtonen P, Kiviniemi O, et al. Risk factors for prolonged inten- sive care unit stay and hospital mortali- ty in acute drug-poisoned patients: an evaluation of the physiologic and labo- ratory parameters on admission. J Crit Care 2011;26:160-5. 3. Lam SM, Lau AC, Yan WW. Over 8 years experience on severe acute poi- soning requiring intensive care in Hong Kong, China. Hum Exp Toxicol 2010;29:757-65. 4. Heyerdahl F, Bjornas MA, Hovda KE, et al. Acute poisonings treated in hospi- tals in Oslo: a one-year prospective study (II): clinical outcome. Clin Toxicol 2008;46:42-9. 5. Lund C, Drottning P, Stiksrud B, et al. A one-year observational study of all hos- pitalized acute poisonings in Oslo: com- plications, treatment and sequelae. Scand J Trauma Resusc Emerg Med 2012;20:49. 6. Teo AI, Cooper JG. The epidemiology and management of adult poisonings admitted to the short-stay ward of a large Scottish emergency department. Scott Med J 2013;58:149-53. 7. Levine MB, Daniel E, Truitt CA, et al. Toxicology in the ICU: Part 1: general overview and approach to treatment. Chest 2011;140:795-806. 8. Mokhlesi B, Leiken JB, Murray P, Corbridge TC. Adult toxicology in crit- ical care: part I: general approach to the intoxicated patient. Chest 2003;123:577-92. 9. Pohjola-Sintonen S, Kivisto KT, Vuori E, et al. Identification of drugs ingested in acute poisoning: correlation of patient history with drug analyses. Ther Drug Monit 2000;22:749-52. 10. Heyerdahl F, Hovda KE, Bjornaas MA, et al. Clinical assessment compared to laboratory screening in acutely poi- soned patients. Hum Exp Toxicol 2008;27:73-9. 11. Montague RE, Grace RF, Lewis JH, Shenfield GM. Urine drug screens in overdose patients do not contribute to immediate clinical management. Ther Drug Monit 2001;23:47-50. 12. Murnion BP, Granot R, Day RO. Utility of urine drug screening: a clinical audit. Emerg Med Australas 2007;19:246-52. 13. Vakkalanka P, Rushton WF, Hardison LS, Bishop MC, Haverstick DM, Holstege CP. Evaluation of the initia- tion of urine drug screens intended for use in transfer patients. Am J Emerg Med 2014 Sep;32(9):1037-1040. 14. Wang GS, Deakyne S, Bajaj L, et al. The limited utility of screening labora- tory tests and electrocardiograms in the management of unintentional asympto- matic pediatric ingestions. J Emerg Med 2013;45:34-38. 15. Persson HE, Sjoberg GK, Haines JA, Pronczuk de Garbino J. Poisoning severity score. Grading of acute poison- ing. J Toxicol Clin Toxicol 1998; 36:205-13. 16. Rockett IRH, Putnam SL, Jia H, Smith GS. Declared and undeclared substance use among emergency department patients: a population-based study. Addiction 2006;101:706-12. Article Non co mmerc ial us e o nly