xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001999363
007 cr mnu|||uuuuu
008 090413s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002475
Khlifi, Abdmalek S.
Etiological characterization of emergency department acute poisoning
h [electronic resource] /
by Abdmalek S. Khlifi.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 98 pages.
Dissertation (Ph.D.)--University of South Florida, 2008.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
ABSTRACT: Poisoning is frequently associated with psychological and physiological co-morbidities that can be assessed in order to improve patients' management and reduce cost. The primary objective of this study is to conduct a review of emergency department (ED) poisonings to characterize its demographics and assess associated co-morbidities. The second objective is to explore correlation between personal history of diseases and poisonings. Predictors for poisonings and its outcomes were investigated and risk factors for suicidal poisoning and how it relates to mental illnesses were explored. Six hundred and forty nine cases admitted to ED between 2004 and 2007 were studied. Results indicate that difference in ethnic background was substantial as poisoning cases were predominantly African Americans (79.9%) between 36-45 years old with a male to female ratio of 1.3.Intentional illicit drug overdose was the greatest risk factor for ED poisonings, and among the 649 cases, heroin overdose was the most common cause of poisoning at 35.4% (n=230), cocaine overdose at 31.7% (n=206), heroin and cocaine overdose at 4.3% (n=28), multiple drug poisoning at 5.5% (n=36), and antidepressant/antipsychotic poisoning at 6% (n=39). A significant correlation between heroin poisonings and asthma (F=20.29, DF=1, p= .0001) was found, as well as between cocaine poisoning and hypertension (F=33.34, DF=1, p=.0001), and cocaine poisoning and cardiovascular diseases (F=35.34, DF=1, p=.0001). Another significant finding is the change in the pattern of the route of illicit drug use from injection to inhalation; it is thought this may reduce the rate of HIV and Hepatitis transmission via hypodermic needles among illicit drug users. As well, inhalation and insufflation may be risk factors that aggravate preexisting asthma.Mental illnesses, chiefly depression, remain one of the greatest risk factors for suicidal poisoning beside age, Hispanic race, gender, ingestion route and unemployment. This study provides supporting evidence that poisoning, particularly deliberate poisoning with illicit drugs remains a serious issue that significantly aggravates co-morbidities and raises treatment cost by increasing both the rate of hospitalization and hospital length of stay (LOS). Pragmatic guidelines and innovations in reducing heroin and cocaine abuse in these patients may lessen the severity of diseases and reduce its burden on the healthcare system and on society.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Advisor: Raymond D. Harbison, Ph.D.
x Environmental & Occupational Health
t USF Electronic Theses and Dissertations.
Etiological Characterization of Emer gency Department Acute Poisoning by Abdmalek S. Khlifi, M.D. A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Environmenta l and Occupational Health College of Public Health University of South Florida Major Professor: Raymond D. Harbison, Ph.D. Rony Francois, Ph.D. Jay Wolfson, Dr. P.H Steve Morris III, M.D. Date of Approval: May 5, 2008 Keywords: asthma, heroin abuse, cocaine abuse, suicide, co morbidity, depression, psychosis, hospital cost, length of stay Copyright 2008 Abdmalek Khlifi
DEDICATION I dedicate this dissertation to my family and to my friends. A special feeling of gratitude to my loving parents, Bacha & Fajrah Khlifi for th eir patience and their sa crifice and support. To my son Skandar and to my Wife, for their patience & support throughout all thes e long years of school To my sisters Dalila, Sabah, Mounjia, Yamina & to my brothers Rchid, Noure ddine Faisal and Houcine. I also dedicate this dissertation in memory of my mother in law who died of a liver failure and never had access to a transplant. To my many friends who have always supporte d me throughout and I will always appreciate all they have done for me. I also dedicate this work to two spec ial friends Dr. Carney & Mr. M. Labidi.
ACKNOWLEDGMENTS I would like to thank a ll the people who helped on this research. First, I wish to thank my advisor, Dr. Raymond Harbison for all his guidance, encouragement, support, and patience. His sincere interests in science and ed ucation have been a gr eat inspiration to me. Also, I would like to thank my committee memb ers Dr. Francois, Rony, Dr. Jay Wolfson and Dr Steve Morris for accepting to be member s of my committee and for their continuous availability, help and support. Additionally, I would like to acknowledge Dr. Leslie Zun who helped and supported me implementing this project. I also would like to acknowledg e all his research tem and fellows that participated gener ously on this project and were ve ry helpful in its completion. I must acknowledge the many friends, coll eagues, students, IMGs, IRB committee membersÂ’ teachers, librarians who assiste d, approved, advised, reviewed, discussed, suggested and supported my research.
i TABLE OF CONTENTS LIST OF TABLES i LIST OF FIGURES ii LIST OF ABBREVI ATIONS iii ABSTRACT iv INTRODUCTION 1 STUDY AIMS AND OBJECTIVES 3 BACKGROUND AND LITERATURE REVIEW 4 STUDY HYPOTHESIS 15 MATERIALS AND METHODS 16 1.0. Design of The Study 16 2.0.Data Collection and Material Studied 17 3.0.Definition of Variables 19 4.0.Data Analysis 21 RESULTS 22 1.0. Demographic Characterization 22 1.1. Age and Gender 22 1.2. Race and Marital Status 24 1.3. Employment and Insurance Coverage 25 1.4. Mode of Arrival 27 1.5. Chief Complaint 28 1.6. Vitals on Admission 29 1.6.1. Temperature on Admission 29 1.6.2. Heart Rate 30 1.6.3. Respiratory Rate 30
ii 1.6.4. Systolic Blood Pressure 31 1.6.5. Diastolic Blood Pressure 31 1.6.6. Glasgow Score 32 2.0. Poisoning Agents 32 2.1. Poisoning and Medical History 34 2.2.Relationship of Poisoning to Smoking and Alcohol Abuse 34 3.0. Poisoning Co-morbidity 34 3.1. Personal Medical History & Diabetes Mellitus 34 3.2. Personal Medical History of Hypertension 36 3.3. PMH of Cardio Vascular Diseases 37 3.4. Personal Medical History of Arthritis 37 3.5. Personal Medical History of Malignancy 37 3.6. Personal Medical History of Asthma 37 3.7. Personal Medical History of Depression 41 3.8. Personal Medical History of Psychosis 43 3.9. Personal Medical History of Other Mental Illnesses 45 3.10. History of Smoking 45 3.11. History of Alcohol 48 3.12. Intake History of Recreational Drug Use 50 4.0. Analysis of ED Management & Final Outcome 51 4.1. Initial Site Management Effect on The Outcome 51 4.2. Poisoning Final Outcome 52 4.3. Correlation of Outcome and Site Management 54 4.4. Prior History of Poisoning 54 4.5. Follow Up 55 4.6. Final Disposition 55 5.0. Analysis of Correlation 58 5.1. Depression Correlations 59 5.2. Psychotic Disorders Correlations 61 5.3. Heroin Poisoning Correlations 61
iii 5.4. Cocaine Poisoning Correlations 62 6.0. Prediction Analysis 64 DISCUSSION 66 CONCLUSION 74 REFERENCES 77 FINANCIAL DISCLOSURE 88 APPENDICES 89 Appendix A: Models Calculation Tables 90 ABOUT THE AUTHOR End Page
iv LIST OF TABLES Table 1 Summary Ta ble Diabetes, Asthma & hypertension 40 Table 2 Summar y list of Depression correlations 60 Table 3 Summar y list of psychosis correlations 61 Table 4 He roin Significant correlations 62 Table 5 Co caine significant correlations 63 Table 6 Model Summary for Poisoning Final Outcome 90 Table 7 Stepwise Regr ession Model for Suicidal predictability 92 Table 8 Heroin poisoning Predictors Model Summary 93 Table 9 Asthma Model Predictors 95 Table 10 Asthma Regression Model Summary 96
v LIST OF FIGURES Figure 1 Po isoning Age Distribution 22 Figure 2 Pois oning Distribution by Gender 23 Figure 3 Distribution of Poisoning by Gender & by Age 24 Figure 4 Pois oning Race Distribution by Age 24 Figure 5 Marital Status Distribution by Age 25 Figure 6 Poisoning by Employment Status 26 Figure 7 Poisoning Employment Distribution by Age 26 Figure 8 Mode of Arrival 27 Figure 9 Chief Complaint 29 Figure 10 Temperature on Admission 29 Figure 11 Pulse on Admission 30 Figure 12 Respir atory Rate on Admission 30 Figure 13 Systolic Blood Pressures on Admission 31 Figure 14 Diastolic Blood Pressure on Admission 31 Figure 15 ED Poisoning Exposure 32 Figure 16 Overall Route of Exposure 33 Figure 17 PMH of Di abetes and Hypertension 35
vi Figure 18 PMH of Hypertensi on and Cardio Vascular Diseases 36 Figure 19 Smoking & Regular Drinking by Asthma Co-morbidity 38 Figure 20 Depression & Psychosis Co-morbidity by age 43 Figure 21 Smoking and Alcohol Co-morbidity by Age 46 Figure 22 History illic it drug abuse by Co-morbidity 50 Figure 23 Major Outcome & Final Disposition by Initial Management Site 52 Figure 24 Admissi on rates by co-morbidity 53 Figure 25 Documented Prior History of Poisoning 54 Figure 26 Percent of Af ter Hospitalization Follow Up 55 Figure 27 Final Disposition Distribution 56 Figure 28 Disposition by ER management time 57
vii LIST OF ABBREVIATIONS LOS: Length of stay CU: Cocaine Users DA: Drug Abuse IDU: Illicit Drug Use HU: Heroin Users HCU: Heroin & Cocaine Users PCP: Primary Care Physician MDU: Multiple Drug Users Binned: Grouped UK: Unknown or missing AA: African American ED: Emergency Department Hisp: Hispanic PMH: Personal Medical History CVD: cardio vascular diseases ED: Emergency Department
viii ETIOLOGICAL CHARACTERIZATION OF EMERGENCY DEPARTEMENT ACUTE POISONING AbdMalek S. Khlifi, M.D. ABSTRACT Poisoning is frequently associated with psychological and physiological comorbidities that can be assessed in order to improve patients management and reduce cost. The primary objective of this study is to conduct a review of emergency department (ED) poisonings to characterize its demogra phics and assess associated co-morbidities. The second objective is to explore correlation between personal history of diseases and poisonings. Predictors for poisonings and its ou tcomes were investigated and risk factors for suicidal poisoning and how it relates to me ntal illnesses were explored. Six hundred and forty nine cases admitted to ED between 2004 and 2007 were studied. Results indicate that difference in ethnic background wa s substantial as poisoning cases were predominantly African Americans (79.9%) betwee n 36-45 years old with a male to female ratio of 1.3. Intentional illicit drug overdose was the greatest risk factor for ED poisonings, and among the 649 cases, heroin overdose wa s the most common cause of poisoning at 35.4% (n=230), cocaine overdose at 31.7% (n=2 06), heroin and coca ine overdose at 4.3% (n=28), multiple drug poisoning at 5.5% (n=36), and antidepressant/antipsychotic poisoning at 6% (n=39). A signi ficant correlation between he roin poisonings and asthma
ix (F=20.29, DF=1, p= .0001) was found, as well as between cocaine poisoning and hypertension (F=33.34, DF=1, p=.0001), and cocaine poisoning and cardiovascular diseases (F=35.34, DF=1, p=.0001). Another si gnificant finding is the change in the pattern of the route of illicit drug use from injection to inhalation; it is thought this may reduce the rate of HIV and Hepatitis tran smission via hypodermic needles among illicit drug users. As well, inhalation and insufflat ion may be risk factors that aggravate preexisting asthma. Mental illn esses, chiefly depression, remain one of the greatest risk factors for suicidal poisoning beside age, Hispanic race, gender, ingestion route and unemployment. This study provides supporting evidence that poisoning, particularly deliberate poisoning with illicit drugs remains a serious issue that significantly aggravates co-morbidities and raises treatment cost by increasing both the rate of hospitalization and hospital length of stay (LOS). Pragmatic guidelines and innovations in reducing heroin and cocaine abuse in these patients may lesse n the severity of diseases and reduce its burden on the healthcare sy stem and on society.
INTRODUCTION Poisonings from intentional substance abus e, as well as from unintentional toxic exposures, remain a significant health concern for hospital emergency departments. It is estimated that there are more than four million incidents of poisoning in the United States each year. The Institute of Medicine has re cently identified that poisonings are the second leading cause of injury-related mortality, with an estimated cost of over $12.6 billion annually ( Koplan et al. 1993 ). The majority of poisonings, particularly minor and moderate cases, are commonly managed at home with the direction of lo cal poison centers via telephone. While the surveillance of moderate poisonings is well documented in this way, severe cases of poisoning that necessitate an emergency department (ED) admission are often not reported to poison control centers, and subseque ntly, the magnitude a nd trends of severe poisonings is unknown, and can only be analyz ed using patient data from hospital emergency departments. This study is inte nded to fill these data gaps, to study the etiological causes of poisoni ngs and to characterize its demographics and trends. The substances responsible for accidental poisoning and the methods used for self-poisoning have changed considerably over time (Howton et al. 1997, Townsend et al. 2001). In many countries, especially in Wester n Europe and North America over the last two decades, the most common agents take n by adults have been illicit drugs (Byard et al. 2004, Townsend et al. 2001 ). 1
The annual incidence of intentional a nd unintentional poisoning in developing countries varies from 0.2 to 9.3 cases per 1,000 population, and continue to increase annually worldwide (Gulsah & al). The characteristics of poisoning in third world countries are slightly different from what is seen in the western world. Agricultural chemicals, particularly pesticides, are still on the top of the list for both intentional and unintentional poisoning and are still serious public health co ncerns in these countries. In the United States, emergency department (ED) visits for attempted suicide and self-inflicted injury are re latively common, serious, and mo st frequent among adolescents and young adults. Self-poisoning is the mo st common method for suicide. The high prevalence of psychiatric and substance abuse disorders suggests there is an association between the two. Knowledge of condition-specific morbidity and mortality for patients with asthma as well as other aggravating factors is essent ial for making critical decisions. Substantive differences have been reported in terms of the prevalence of drug abuse among asthma patients; however the etiology and the mechanis ms of action for this association have not been explained. There is a need, especially for the unde rserved minority communities that have high frequencies for certain methods of drug us e, to understand the factors involved in poisoning and how to effectively design prev ention programs for the purpose of reducing their occurrence and adverse outcomes. 2
STUDY AIMS AND OBJECTIVES The objectives of the study are: To c onduct a retrospective chart review to characterize the demographics of acute poisoning of patients that visited or were admitted to ED (ED poisoning) between December 2004 and May 2007; To identify the most common substances associated with poisonings and stratify them by route, age, gender, time of the day and reason for admission. Co-m orbidities of mental disorders, asthma, diabetes, chronic hypertension, cardiovascular diseases, arthritis, cancer, drug abuse and occupational exposures will also be examined to assess their correla tion with ED treated poisoning. A cost assessment was also designed to assess the burden of heroin and cocaine poisoning on asthma patients. Specifically this study will assess the type and route of poisoning and stratify them by illicit drug, pharmaceutical, chemical, occupational, industrial or domestic poisonings. This inves tigation will assess the circumstances of the poisoning including accidental or intentional suicidal. The ra te of hospitalization and the final outcome of poisoning events will be examined. This study will evaluate the rates of complications from poisonings as well as the le ngth of stay and the casualty rates if any have occurred. On site initial management r eceived by patients at home or administered by patients families before arrival to ED and its effect on the final outcome of the case will be examined in order to estimate which of these factors are most predictive of the outcome and poisoning final disposition. 3
BACKGROUND AND LITERATURE REVIEW Poison exposure is a significant health concern and it is the third most common cause of death that occurs at home (Schauben & Spillane, 1991) While the overall true incidence is unknown, it is estimated by many res earchers that there are over four million poison exposures per year (Litovitz et al. 1995.) According to the Center for Disease C ontrol and Prevention (CDC), cases of poisoning are grossly underestimated and a significant number of th em never reach the hospital since many of the minor case are tr eated at the patients home and are never recorded through poison control centers. Among all deaths due to poisoning, carbon monoxide intoxication is the most prevalent. Over 86 percent of all reporte d poisoning cases were relate d to accidental exposure, while suicidal or self poisoning accounted fo r less than 8% of the total reported poisonings (Litovitz et al. 1995; Simonsen, 1992) The largest proportion of accidental poisonings represented events that were preventable. The Healthy People 2000 report has called for the importance of the reduc tion, prevention, and tr eatment of poisoning (Litovitz et al. 1994). In a study conducted in Barcelona, Spain, it was estimated that poisoning constituted 5.2% of all emergencies. Am ong poisoning cases reviewed in this study 87.2% were due to heroin related events, 13.1% were due to acute alcohol intoxication and 6.2% were due to acute medicinal intoxi cation. During the study period of 3 years, the authors identified over 233 emergencies involving cocaine. Of these, 55% were 4
female with a mean age of 28.3 years. Th e authors also reviewed 1,134 patients with episodes of acute poisonings seemingly related to pharmaceutical drugs. The most commonly used drug was benzodiazepines at 56.7%, followed by analgesics at 12.6% and barbiturates at 11.7%. The Barcelona study also identified 1,456 toxicological emergencies; 62% of these were suicidal a ttempts, and 12.6% accidental intoxications. It is important to note that in this study the author has include d withdrawal syndrome within the definition of heroin and cocai ne poisoning in his statistics (Monteis Catot et al. 1990). Drug abusers are more likely to be hospita lized, and 2.3 times more likely to use an emergency room than non-abusers (Stein et al. 1993). Patients using drugs are frequently seen in EDs and drugs encountered generally reflect regional and city variation, as drug obsession sweeps through subcultures. It is now known that there is a tendency for specific drug subcultures to grow locally around the use of specific agents. The overall local drug use trends in the United States i ndicate that cocaine a buse decreased between 1992 and 2002 (Bernstein et al. 2005). On the ot her hand cocaine abuse decreased in five states by more than 60% it increased 100% or more in four other states. A considerable part of emergency care is treating medical complications of substance abuse, but the act ual abuse of drugs is of ten not addressed in ED (Bernstein et al. 2005). This lack of interest to identify drug abuse may relate to the fact that physicians, particularly emergency physicians (EPs), use illicit drugs as often as the general population (Hughes et al. 1999). Marijuana is the most commonly used drug in the United States, but leads to less ED visits than other drugs. Researchers es timated that there were over 2.6 million new marijuana users in the year of 2001. Over 3.1 million Americans use marijuana daily, and 5
over 14 million of Americans aged 12 and older use marijuana at least once a month (NIDA 2008). In recent years marijuana has beco me more potent than of what it was in the 1970s and 1980s, which resulted in mo re ED visits among youth, and in 2002, marijuana became the third most common drug seen in ED. Marijuana use increased by over than 24% between 2000 and 2002 ( Whalen et al. 2006, NIDA 2008) Another significant increase seen in recent years is th e use of non-crack cocaine among tenth graders, which has risen from 1.1% in 2003 to 1.5% in 2004. The same study, however, reported that heroin, crack cocaine, ha llucinogens, phencyclidine amphetamine, tranquilizers, sedatives, and methaqualone use remained stable among youth from 2003 to 2004 ( NIDA, 2006). Heroin use remains to be a serious problem in some regions of the United States. A recent change in the route of use of heroin was noted, and there is a new practice of mixing he roin and heating it in aluminum foil then inhaling through a straw, or mixing it with co caine promoting the significant shift from heroin injection to heroin i nhalation. Heroin abuse is asso ciated with fatal overdose, spontaneous abortion, and serious infectious diseases like HIV and hepatitis among IV abusers, and rare cases of spongiform leukoencephalopathy. Although heroin, cocaine, and marijuana remain the most frequently abused drugs leading to ED visits, ED visits involving na rcotic analgesics increased 153% from 1995 to 2002. The use of drug combinations were involved in 75% of the drug abuserelated ED visits, and dependence was the underlying factor in these ED visits. In 2003, about 4% and 4.5% of high school seniors reporte d nonmedical use of oxycodone. These users have a higher economic status. (D'Onofrio Gail et al, 2006). 6
The Substance Abuse and Mental Health Se rvices Administration estimates that in 2003, 6.3 million Americans abused prescription dr ugs. Most abused pain relievers (4.7 million); others abused tranquilizers (1.8 millio n), stimulants (1.2 million), and sedatives (0.3 million). Nonmedical use of oxycodone in creased from 11.8 million users in 2002 to 13.7 million users in 2003. (Volkow, NIDA 2004/2005). A shift to smoking has occurred for the use of stimulants and this route has increased from 12% in 1992 to over 50% in 2002. In 2003, an estimated 378 thousand drug users met the diagnostic criteria for stimulant dependence in the United States ( NIDA, 2006). Cocaine is a strong central nervous system stimulant. It can result in acute myocardial infarction or stroke and sudden death. Cardiac co mplications, such as cardiac arrest and seizures are generally the main causes of cocaine-related deaths. The use of cocaine occurs in a binge pattern, during which the drug is repeatedly taken at high doses. If cocaine is mixed with alcohol, the product is cocaethylene intensifies cocaines euphoric effects, but also increases the risk of sudden death. The want ed effect of cocaine is usually the resulting euphoria that lasts 15 to 20 minutes if snorted and 5 to 10 minutes when smoked. In addition to euphoria, cocaine can raise blood pressure, cause tachycardia, constrict vess els, and dilate pupils. The wanted effect of heroin users is euphoria, (also called rush); heroin can also cause flushing of the skin, and dry mouth. When tolerance develops, heroin users become obliged to use higher doses in order to achieve the same wanted effect. Opioid overdose causes respiratory depr ession that can be reversed by Naloxone. The onset, severity, and duration of withdr awal symptoms depend on the opioid abused 7
and the extent of dependence. Sixty-two per cent of primary heroin admissions reported injection as the route of administration, 33% reported inhalation, and 3% reported smoking. Maintenance treatment for heroin addi cts involves daily or al synthetic opiate administration usually methadone. Buprenorphine is also a medication now available for treating addiction to opiates. Treatment of teenagers for marijuana dependence and abuse has increased 142% since 1992. Marijuana is known as a recreational drug, particul arly to teens. Teenagers are three times more likely to be treated for marijuana than for alcohol and six times more likely to be in treatment for marijuan a than for all other drugs combined (Stephens et al. 2000). The proportion of admissions for abuse of stimulants increased from 1% to 7% between 1992 and 2002. Abuse of prescription me dications has increased in all segments of the population. In 2003, about 4% to 4.5% of high school seniors reported nonmedical use of oxycodone (NIDA, 2006). Club drugs are sometimes preferred over other types of drugs like marijuana and heroin or methamphetamine because some users see them as social interaction enhancers. Club drugs such as GHB, flunitrazepam or ke tamine are taken in combination, or with alcohol. They are easy to acquire and they ar e generally favored over marijuana, lysergic acid diethylamide, methamphetamine, and opiates (Gahlinger, 2004). Hallucinogens are less commonly used. Most users of lysergic acid diethylamide voluntarily decrease or stop its use over time. Lysergic acid diethylamide is not considered addictive (NIDA, 2006 ). 8
Gulsah et al. 2005 reviewed the frequency a nd mortality risk fact ors of acute adult poisoning in Adana, Turkey from 1997 to 2002 and included 2,229 adult patients. The authors reported that 1.6% of their emergency cases were poisoned and the majority of these poisoned cases were attempted suicid es (76.4%). Drugs were the most common (59.0%) poison used, and pesticides were the second (26.4%) most common substance encountered. Among drugs, psychoactiv e were the most common (35.5 %). The number of drug abusers is increasing across the U.S. at an alarming rate, though many of the related deaths are prev entable. According to a recent report conducted in Multnomah County, Oregon, hero in alone was implicated in over 3805 deaths between 1993 and 1999 (Koplan et al., 2000). Heroin has greatly increased in popularit y in recent years. For many young people, it is now the drug of choice, with a dose or "hit" available on some streets for the same price as a pint of beer. The low price indicates there is a signif icant supply of the drug. Low and middle class youth seem to be particularly lured by the drug; and the most prevalent route of administration has also shifted since the early 1990s to sniffing and snorting over injecting the drug. The drug has been widely avai lable, and heroin on th e street now is of a higher purity and better quality than before. The association of drug abuse co-morbidit y, particularly cocaine and asthma, has been reported in few articles. A study c onducted in Bronx, NY, where the author studied the occurrence of new onset bronchospasm or recrudescence of asthma, established a correlation between cocaine use and new onset asthma. ( Osborn & Al. 1997). 9
The evidence of heroins effect on the resp iratory system, particularly the lungs, has been studied by Steensen et al in 1993 who established a direct relationship between heroin abuse and pulmonary edema. This rese arch did not clearly describe the mechanism of action but it was concluded that one possibility is the increased pulmonary permeability of the lung capillaries in res ponse to the exposure. In 2004 Marby et al found a pattern of heroin overdos e-induced pulmonary edema. In the early twenty first century, cocain e and heroin abuse was studied, and were found to increase the rate of intubations and hospital utilization in patients with acute asthma exacerbation (Levine et al. 2005) When heroin was first discovered to be corr elated to patients with life threatening asthma in the early 1990s from a limited set of cases, it was initially thought that this was a novel phenomenon that did not exist before, and was related to the change in the pattern of the drug exposure from injection to inhalation (Krantz et al, 2003). For the purpose of verifying if this is a new phenomenon, a limited retrospective study was conducted which concluded that this relationship has existed prior to the 1990s. In reviewing the literature however, it was found that this model of correlation was also described in 1988 when an unexplained exacerbation of asthma in 3 cases that required intubations was linked to heroin use (Hughes et al.) Heroin insufflation was then studied by Krantz at al. 2002 to further investigate this theory and attempted to demonstrate that the insufflation of he roin functioned as a trigger for patients with life th reatening asthma. In this study, the authors reported that of the total 104 patients admitted to ICU, most patients were not significantly different in mean age or sex between groups of patient s with diabetic ketoacidosis (DKA) and 10
asthma. Although the racial distribution di d differ in that the DKA group had more Hispanics than whites in th e asthma group. However compared to ICU admissions with DKA, asthmatic patients reported higher pr evalence of heroin use (41.3% vs. 12.5%). Asthmatics were also more likely to report cocaine use, but the difference between the two groups was not significant. Levine et al. identified 152 ICU patients of which he found 27.6% used cocaine without heroin (and they were classified as cocaine users) and 30.9 % used heroin with or without cocaine (and were classified heroin users). Cocaine users ha d longer lengths of stay than non-users (3.4 vs, 2.5 days p<0.049) and were more frequently hospitalized and intubated. Heroin users were intubated more frequently than non-users. Neither hospital stay nor percentage of IC U admissions was significant (Levine et al. 2005). The author showed that both cocaine and heroin mi ght have induced asthma exacerbation by multiple effects on the bronchus, such as th e morphine bronchoconstrictive effect (a metabolite of heroin through deregulation of mast cells and subsequent release of inflammatory mediators particularly histamine) (Withington et al. 1993). In a case series reported in late 2005, Cygan et al. de scribed 5 cases of status asthmatics that were induced by heroin inha lation. Though he hypothesized that the cases were directly triggered by the inhalation of the drug, the author was unable to establish a single mechanism of how he roin triggered asthma. There are several suggestions about the mechanism of action of heroin related asthma effects, but pulmonary mast cell degr anulation secondary to heroin may be very plausible as indicated by an experiment that used intradermal and IV morphine and synthetic opiate administration, resulting in a stimulation of mast cells to release 11
histamine. This release of histamine is a direct pharmacologic mechanism independent of opiate receptors. However, none of the studies that suggest heroin acts as a trigger for asthma have reported that the exacerbation occu rred at the time of the first use. This makes the histamine release theory very unlikely though it seems very appealing. Unless the heroin sensitization period is extremely long, which is contrary to what is currently known about immunological sensit ization. Association of other factors that are receptor independent might be also possible. The controversy regarding this issue remains, given the fact that opioid receptor action is thought to result in reduction of bronchoconstriction by blocking the cholinergic f actor in bronchoconstriction. Asthma is a multifactorial disease influenced by genetic and environmental factors (Matthis et al. 2006). A recent study have revealed the involvement of some receptors regulated by inherited genes in the pathogenesis of asthma that could predispose some families or ethnic groups to higher risk for acquiring asthma (Lachheb et al. 2008). In the past decade, several loci and over 1100 genes have been found to be associated with the disease in some populations. Among these lo ci, region 12q13-24 has been implicated in asthma etiology in multiple populations, suggesting that it harbors one or more asthma susceptibility genes (Balaci et al. 2007). On the other hand there are significant and distinct changes in th e profile of gene expression in the human nucleus associated with cocaine and heroin abuse (Albertson et al. 2006) and several gene variants have been also detected and inferred to cont ribute to heroin abuse and a ddiction. For most of these genes, however, the relationship to the pa thophysiology of asthma remains speculative, and none appear to be direct ly involved in th e activation of airw ay inflammatory 12
processes or allergies, or explain fully a ll the aspects of the complex mechanism of addiction and abuse of heroin. Several studies suggest the presence of a re lationship between me ntal illnesses and poisoning for self-harm. Attempted suicide is known to be risk factor for committed suicide, and certain characteristics were se en among those who commit suicides such as having made previous suicide attempts, or having a family member who has attempted suicide. One of the studies has shown that a ttempted suicide is most closely related to depression. But the study concl uded that patients with depre ssion were significantly more often aged, male, married, and used methods other than poisoning by solids or liquids compared with patients with other mental disorders (Ichimura et al. 2005). Depression is the disease most often implicat ed in suicide. In many countries, there has been a consistent increase in suicides among adolescents and young adults, as well as a decrease in suicides among women. Th e management of depression includes antidepressant medication. However, appropriate antidepressant prescribing is a very important factor in terms of preventing fatal antidepressant overdose. The actual risk of suicide in patients prescribed an antidepre ssant may be up to 10 times greater than for others. The likelihood for serious and poten tially fatal poisoning following overdose differs considerably among the drugs avai lable. Most deaths from overdose are attributable to tric yclic antidepressants (Henry 1996). Howton, in a meta-analysis focused on patie nts with bipolar disorder in which suicide (13 studies) or attempted suicide (23 studies) was reported as an outcome. The study identified the main risk factors for suicide as previous suicide attempt and hopelessness. The main risk factors for nonfatal suicidal behavior included family history 13
of suicide, early onset of bipolar disorder extent of depressive symptoms, increasing severity of affective episodes, the presence of mixed affec tive states, rapid cycling, comorbid Axis I disorders, and abuse of alc ohol or drugs. The author concluded that in patients with bipolar disorders, prevention should include atten tion to these risk factors in assessment and treatment, including when d eciding whether to initiate treatment aimed specifically at reducing suicide risk (Hawton, et al. 2005). According to Nordentoft et al. 2007, who focused their work on the prevention of suicide in Denmark, where suicide rates have been declining in the last two decades for all age groups, the decline seems to be larg est among females. Their conclusion is that the decline is mainly due to the focus on ps ychiatric patients, esp ecially schizophrenic related disorders and homeless persons. In his study he identified that the most significant risk factors for suicide are: male gender, young age, short duration of illness, multiple admissions during the prior year shorter time since discharge from a healthcare facility, co-morbid depression, drug abuse, poor compliance with medication and high IQ (Nordentoft et al. 2007). Another study conducted in Mexico in 1999 co ncluded that there were high rates of suicide related depressive disorders in a predominantly female population. Drug products, according to the author, were more frequently used than home or work related chemicals. The mortality related to poison exposure wa s very high at 27% and 16% with permanent squeal, though advanced lif e support and antidote medications were available (Kingsbury et al. 1999). 14
STUDY HYPOTHESES 1Drug abuse overdose is the gr eatest risk factor for ED poisoning compared to the other environmental, occupationa l or prescription medication drugs. 2There is a correlation between preexisting history of chronic disease, such as asthma and intentional poisoning using heroin or cocaine. 3Mental illness, such as depression and psychotic disorders, are risk factors for ED poisoning. 4Illicit drug poisoning is a burden on asth ma patients, and it incurs longer hospital stays and higher hospital costs compared to those who do not present with a history of illicit drug poisoning. 15
METHODS AND MATERIAL 1.0DESIGN OF THE STUDY The study population consists of Chicago city residents visiting the Emergency Department (ED) of the Sinai Hospita l between December 2004 and May 2007. The Sinai System provides virtually all-medical and surgical ca re in the area and it also includes a childrens hospital a nd a rehabilitation center. The hospital is serves the population 5 miles west of down town. This population is diversified between African American, Hispanic, white, and Indian ethnicities. This study is retrospective, and all cases of poisonings were identified through a search of the electronic records using the In ternational Classification of Disease Ninth Revision, Clinical Modification (ICD-9-CM). A database of all ED poisoning cases was compiled from ED electronic records. Data pe rtaining to history of chronic disease comorbidities, as well as to poisoning by illic it drugs, pharmaceutical, and chemical agents was collected. The poisoning exposure variable descri bed the type of poison, the poisoning exposure route, and the circumstances of int oxications; (intentional, unintentional or selfpoisoning or attempted suicide). Initial manage ment of the event was also collected to determine its adequacy and to correlate it to the final outcome and final disposition of the poisoning cases. 16
Data on the history of prior poisoning was collected in addition to smoking, alcohol, and illicit drug use. The data collect ion preprinted sheet in cluded also personal history of mental illness such as depression, psychosis and anxiety disorders. Data on the initial management received and follow-up, as well as the outcome of the event as assessed by the treating physician at the time of transfer or discharge was collected. In order to assess hospital cost and the burden of drug abuse on asthma, a follow up study was designed to collect co st data and assess expenses, including expenses paid by insurance and by self-payers. Two groups of pa tients were selected : the first group of 137 patients was selected from the 649 patients in itially compiled for poisoning. The second group was randomly selected from the pool of asthmatics that were not known to have any abuse history or hospitalization for any dr ug related event. Age, race, and insurance coverage data was collected as well as hos pital charges, and amounts collected, expenses, and length of stay for those hospitalized. 2.0. DATA COLLECTION & MATERIAL STUDIED The data collection was performed acco rding to HIPAA regulations and after approval by the hospitals Institutional Re view Board (IRB) committee. The system database was queried for cases of pois oning during the period between December 2004 and May 2007 to generate a list of patients to be enrolled into the study. Data abstracted from patients medical re cords included age, gender, race, date and time of the visit, marital status, employme nt, insurance coverage, chief complaint and vitals on admission. Information on the patien ts personal medical hi story of diabetes, 17
mellitus, hypertension, cardio-vascular diseases asthma, arthritis, HIV, hepatitis and seizure disorders was recorded. Initial medical management and assessment of the gravity of the case was recorded and then weighed against the final outcome in order to discern the cases that were not considered serious, but resulted in an adverse outcome; this may help ED determine areas to implement preventive m easures in order to avoid any possible underassessment of patients at risk. Any discre pancy in time of initiation of the treatment after patient arrival was recorded and delays will be compared and correlated to the final outcome of the patient. This includes the cas es that were managed in a nonhealth care facility or at the site of e xposure. The percentages of patient s initially treated and referred from another health care facili ty have also been evaluated. Treatments, such as the use of antidotes, were recorded to assess the availability of necessary treatment as well as any de viation from the recommended standards. The management provided was then classified in the following categories: Decontamination only, Observation only, No therapy provi ded, Decontamination and other therapy, other therapy only (without decontamin ation). Patients health coverage, unemployment, and history of me ntal illness such as depres sion, psychotic disorders, or personality disorders were an alyzed. Significance of ment al illness in the group of patients who repeatedly attempted drug or ch emical intentional se lf poisoning has been explored and the correlation of addiction hi story with mental illness co-morbidity has been examined. Voluntary vs. unintentional, domestic chemi cal use vs. agricultural and industrial, occupational vs. environmental, pharmaceutical intentional vs. pharmaceutical adverse 18
reaction illicit drug use, and recreational intoxication have been stratified by age, sex, socioeconomic status, employment, presence of history of chronic illness, alcohol use and mental illness have been examined for correlation with the final outcome of the poisoning cases. Although morbidity and mortality of acute poisoning are increased in elderly compared to younger patients, little has been published on this issue. To investigate the influence of age on the clinical course of acute poisoning, the cause of acute poisoning and the clinical outcome have been recorded. As compared with younger patients, mean length of stay at the ICU, if found to be significantly higher this might indicate a more serious course of acute poisoning. 3.0. DEFINITION OF VARIABLES Poisoning was defined as a condition or physic al state produced by the ingestion, injection or inhalation of, or exposure to a deleterious agent. Poison exposure may include any medicinal drug, environmental, agricultural, industrial, or occupational substance that may or may not result in a visit to the Emergency Department (ED). Environmental poisoning is any passive, non-occupational exposure that results from contamination of air, water, or soil. Occupational exposure is an exposure that occurs as a direct result of the workplace. Therapeutic error is an unintentional deviation fro m a proper therapeutic regimen that results in the wrong dose, incorrect route of administration, administration to the wrong person, or administration of the wrong substance. Drug interactions resulting from unintenti onal administration of drugs or foods which are known to interact are also included. 19
Unintentional misuse differs from intentional misuse in that the exposure was unplanned or not foreseen by the patient. Bite/sting includes all animal bites with or without stings. Food poisoning is a suspected or confirmed inge stion of food contaminated with microorganisms. Suicidal poisoning is a poisoning that results from an inappropriate use of a substance for reasons that are suspected to be self-destructive. Intentional misuse is defined as an exposure resulting from the intentional improper or incorrect use of a substance for reasons ot her than the pursuit of a psychotropic or euphoric effect. Unintentional poisoning is all unintentional exposures not otherwise defined. Intentional abuse is defined as an exposure resulti ng from the intentional improper or incorrect use of a substance where the victim was likely at tempting to achieve a euphoric or psychotropic effect. All recreational use of substances for any effect is included. The category malicious poisoning is used to capture patients who are victims of another persons intent to harm them. Minor effect is a situation where the patient deve loped some signs or symptoms as a result of the exposure, but they were mini mally bothersome and resolved rapidly with no residual disability or disfigurement. Moderate effect occur when patient exhib its signs or symptoms as a result of the exposure that were more pronounced, more prolonged, or more systemic in nature than minor symptoms. Some form of treat ment is indicated but Sympto ms were not life threatening, and the patient had no residual disability or disfigurement 20
Major effect is defined as when patient exhibits signs or symptoms as a result of the exposure that was life-threaten ing or resulted in a hospita lization or in ICU with or without significant residual disability or disfigurement Death occurs when patient dies as a result of the exposure or as a direct complication of the exposure. 4.0. DATA ANALYSIS Data collected was reviewed and checke d for completeness before being entered into a database and missing values were excl uded using a pairwise deletion. The data was then analyzed using SPSS 16.0 statistical packag e. Continuous variables such as age, duration of hospital stay, and duration of time between poiso ning and hospital arrival was analyzed using students t test or one-way analysis of variance (ANOVA) between groups. The categorical data be tween groups was analyzed by using the Chi square test; Odds Ratio, and confidence interval, (CI) wa s calculated by a full factorial univariate analyses and stepwise multiple logistic regression analysis. Correlation analysis was conducted using the Pearson r correlation coeffi cient. Results obtained will be presented as mean SD and percent (%). A p value of less than 0.05 will be considered significant for the various association between variables specifically association between asthma and exposure, between exposure and outcome, fi nal disposition, suicidal poisoning and chronic diseases such as depression psychosis 21
RESULTS Between December 2004 and May 2007 a tota l of 649 patients encounters were identified. The poisoning diagnosis was ma de based on a documented acknowledgement by the patient for taking the drug or by a positive urine drug screen that is done systematically for any suspect ed specific exposure. The fi rst section will present the demographics of the exposure, the second section will explore poisoning agents of exposure, the third will describe co-morbidi ties, the fourth sec tion will describe the correlations between variables, and the fi fth section will describe predictors. 1.0. DEMOGRAPHIC CHARACTERIZATION 1.1Age and Gender The distribution of age and gender is shown in Table 1 and Figure 1; age varied between 1 and 89 years with a median of 39.7 years and a standard deviation of 13.6. FIGURE 1: Age distribution: cases were predom inantly adult between 36 and 45 years old. Pediatric cases younger than 15 years old represented only 4.6%. 22
In the pediatric population 4.6 % were younge r than 15 years old and, as shown in Figure 1, the most frequently affected ag e was 48 years old and the most commonly affected age group was between 36 and 45 y ears old. The middle aged adult population for both sexes represented 35.7% of the total cases. The younger adult group of 16 to 35 years old represented only 10.6% and ranked th ird after the 46-55 years old that scored 22.7%. Of the total of 649 poisoning cases th ere were more male, than female, with 42.7% (n=277) females and 55.9% (n=363) male to female ratio of 1.3. FIGURE 2: Poisoning distribution by gender: more male, than females were seen with male to female ratio of 1.3. Below (Figure 3) is a side-to-side (or a mirror effect) graph of the male to female frequency on an age range. This population pyramid allows a better comparison of the various gender differences by age brackets. The distributions of gender by age seem globally similar, but clearly the peaks look di fferent where females are more common at ages 40 to 43 as compared to males that peak at 46 to 49 years old. Among children it 23
seems there were more poisoning cases among male infants at very early age (2-3 years) than females. FIGURE 3: Population Pyramid: distribution of poisoning cases by gender and by age. Poisoning population peaks at 49 years old in male s and at 40 to 43 in females. 1.2Race and Marital Status The difference in ethnic background was s ubstantial as the poisoning cases were predominantly African Americans (79.9%). FIGURE 4: Poisoning distribution by race and age: The poisoning population race is predominantly African American then Hispani cs and Whites in third position. (Race_AA: African American, Race_Hisp: Hispanic) 24
Hispanics were the second most preval ent with 11.9%, followed by Whites at 6.6%. Over 69.3% (n=450) of the total case s were single in comparison to only 12.3% (n=80) were married. Results of the married and unmarried patients by age are displayed in Figure 5. FIGURE 5: Marital Status. Distribution of married versus unmarried patients shows a predominantly single population as compared to only less than 20% married (Marit_Single: Marital status single; Marit_Marr: Married) There were more singles cases than married cases and singles cases were predominantly aged between 37 and 49 years old. 1.3Employment and Insurance Coverage Most patients were se lf-covered 45.8% (n=297), 35.1% (n=228) of the cases were covered by the Illinois Department of Public Aid (IDPA). Only 5.9% (n=38) patients had private coverage. Medicare a nd Medicaid covered only 8.2% of the total cases. 25
FIGURE 6: Poisoning by employment status. Most poisoning cases were unemployed and not covered by any insurance, shown in Figure 6 and Figure 7. Most patient with poisoning episodes brought to the hospital were unemployed 87.4% (n= 567) as opposed to only 9.4% (n=62) of employed. FIGURE 7: Population pyramid, Employment distribution by age. The age group that is the mostly affected was between 37 and 52 years old. 26
1.4Mode of Arrival Fifty three percent (n=345) of patie nts with poisoning were brought by the Chicago fire department in response to a 911 call from a patients families, or from the police department. Twenty nine point six percent of patients (n=192) walked in the emergency department or were brought by a family member or in a wheelchair. Only 0.5% of the case visited their primary care physician (PCP) before presenting to the emergency department. FIGURE 8 : Mode of arrival of patients with poi soning to ED. Chicago Fire Department brought more than half of the total cases. The Pearson correlation test showed that there wa s a significant correlation between mode of arrival and age (r =-.119, DF=589, p=.004), marital status (r=.113, p=.006 and DF=627), pulse (r=-.111, DF= 622, p=.007), exposure (r==.134, DF=591, p=.001) and also with the final dis position (r=.111, DF=582, p=.007). Though these correlations are not strong, it could reflect th at the predominance of heroin and cocaine cases mostly brought by police and fire depa rtment resulted in mo re hospitalizations compared to the other patients that just walk ed in or were brought by other means. Also, these correlations could be rela ted to the fact that the outco me is generally worse when 27
EMS brings patients given that EMS is generally called only for situations that seemed severe or had potential for complications. Arrival mode was also significantly correlated with the final disposition particularly to hospital admi ssion (p=0.04). No correlation was established between death and the way the patient arrived to ED. 1.5Chief Complaint Analysis Given the wide variation of the diverse symptoms related to poisoning, the chief complaint was classified in seven categories th at were estimated to be the most common. Category one includes unresponsiveness, Loss of consciousness and or Coma and it included 22% (n=143) of the total patients. Category two includes cardio vascular complaints such as chest pain, angina, and or coronary artery disease. This category was the most re presentive in our series and more than 25.9% (n=168) of the cases presented with these chief complaints. Category three is pulmonary related complain ts such as Shortness of breath (SOB), cough, wheezing or exacerbation of a preexisti ng asthma. In our series, 22.2% (n= 144) of patients presented with this latter comp laint. Category four is drug overdose related symptoms, this category involved 15. 4 % (n= 100) of our patients. 28
FIGURE 9: Chief complaint at the level of ED triage. Most patients presented for chest pain or were brought by CFD unresponsiv e or with shortn ess of breath. 1.6Vitals on Admission 1.6.1Temperature on Admission FIGURE 10: Admission temperature ranged between 86.0 and 105.0 with a median of 98.0 degree Frenheight. 29
1.6.2Heart Rates As shown below in Figure 12, only 2.6% presented with a bradycardia less than 50 and 56.4% (n=366) presented with pulse between 50 and 100 and 39.4% (n=249) with pulse over 100. FIGURE 11: Pulse at arrival to ED. over 35 % of patient had a tachycardia that was higher than 100beat per min. 1.6.3Respiratory Rate FIGURE 12: Respiratory rate on admission. RR was normally distribute d that peaks in the high 20s 30
1.6.4Blood Pressure Systolic FIGURE 13: Frequency and percent systolic blood pressure on arrival to ED. Over 30% of patients presented with systolic high BP over 140 and 10% presented with a systolic BP over 200mm/hg. 1.6.5Blood Pressure Diastolic Diastolic blood pressure was not significantly correlated with either heroin or cocaine use, but it was correlated to alcohol use (p= 0.015) analgesics (p= 0.24) and to marijuana use (p=. 006). FIGURE 14: Frequency and Percent of diastolic Blood Pressure on arrival to ED 31
32 1.6.6Glasgow Scores Glasgow coma scale, known as Glasgow sc ore; a reliable tool for assessing consciousness was significantly correlated to heroin (p=0.002) and to cocaine overdose (p=0.002) 2.0. POISONING AGENTS Poisoning exposure was grouped into fifteen toxic substances or drugs and in four different categories: pharmaceutical or medici nal drugs, recreational drugs, chemicals or chemical industrial. Circumstances of the exposure were also captured and categorized into seven intent groups that included: suicid al, abuse, intentional, misuse, unintentional and therapeutic and adverse drug events (ADE ); the route of exposure was recorded as: inhalation, ingestion, and injection, dermatological, vaginal and or rectal. FIGURE 15: ED poison exposure. The poisoning exposure most commonly encount was heroin followed by cocaine and antidepressant and antipsychotic medications. Multiple drugs inta ered kes were the fourth most commonly seen poisoning exposure in this study population.
33 gs, a, multiple drugs, unknown pills, cardio vascular drugs, anti-cold drugs and others ons gestion in te of exposure was the most commonly seen route particularly for cocaine and heroin. The fifteen exposure substances iden tified and most commonly encountered in ED included heroin, cocaine, alcohol, heroin concomitantly with cocaine, multiple dru ASA (aspirin), anti-depressors and antipsych otics, stimulants, analgesics, methadone, marijuan In summary agents of poisonings encount ered was as follow: heroin represented 35.4% (n=230), cocaine 31.7% (n=206), heroin and cocaine concomitantly taken 4.3% (n=28), alcohol 2.2% (n=14), multiple drugs 5.5% (n=36), anti depressant antipsychotic 6% (n=39), ASA 3.4% (n=22), cardio vascular drugs 2% (n=13), an ti-cold medicati 1.1% (n=2.2), methadone .6% (n=4), CNS stimulants .8% (n=5), Analgesics 1.5% (n=10), Marijuana .5% (n=3), unknown pills 1.7 (n=11), other substances 2.9% (n=19). The route of the poisoning exposure include d inhalation 61.5% (n=400), in 21.1% (n=137), rectal vaginal 10.6(n=69) and injection at 5.2% (n=34.). FIGURE 16: Overall route of poisoning exposure. Inhalation as rou
The exposure circumstances included: abuse 76% (n=498), suicidal 13.4% (n=87), misuse 2.2% (n=14), unintentional 3.2% (n=21), therapeu tic 2.5% (n=16) and ADE .6% (n=4). 2.1Poisoning and Medical History There was a significant relationship between asthma and the exposure to heroin (F=20.4, DF=1, p=0.0001) as well as between hi story of cardio vascular disease and exposure to cocaine (F=35.34, DF=1, p=0.0001). Another significant relationship was revealed between the use of cocaine and history of preexisting hypertension (F=33.34, DF=1, p=0.0001). Depression and Psychosis we re significantly a ssociated with the overall poisoning exposure variable with (F=42.59, DF=1, p= .0001) and (F=6.96, DF=1, p=.0001) respectively. 2.2Relationship of Poisoning to Smoking and Alcoholism A set of significant relationships were established between poisoning exposure and the history of smoking (F=2.90, DF=14, p=0.0001). Another significant relationship (F=3.76, DF=14, p=0.0001) was established between the history of alcohol drinking and the exposure to poison. History of recrea tional drug use was also found significantly related (F=33.64, DF=14, p=0.0001) to th e overall exposure variable. 3.0. POISONING CO-MORBIDITY 3.1. Personal Medical Histor y of Diabetes mellitus There were over 93 diabetic cases that visited the emergency department for poisoning during the study period and they represented 14.3 %. Twelve cases were accounted unknown as the history of prior di agnosis was not clear from the patients records. 34
FIGURE 17: Personal medical history of diab etes and hypertension by age. The distribution of the cases of diabetes looks uniformly distributed and it reflected the distribution of the poisoning exposure within the total popul ation (on the Y access Sum refers to frequency). The diabetic patients ages ranged between 1 and 89 years with a mean of 46.5 and a standard deviation of 11.96, similar to the overall age distri bution of the population studied. More than 59.1% (n=55) were male, 44.1% (n=41) used exclusively cocaine and 29% (n=27) used exclusively heroin, 2.2% used a combination of both heroin and cocaine, and 6.5% (n=6) combined multiple drugs. Among poisoned diabetics patients there were 59.1% (n=55) who also had an associated hypertension and 35.5% (n=33) with cardio vascular diseases and 48.4% (n=45) of asthmatics. Depression was prevalent among diabetics w ith 21.5% (n=20), no correlation with the overall poisoning exposure variable was established. 35
3.2Personal Medical History of Hypertension Over 31% (n=199) cases were diagnos ed with prior history of hypertension. Among these there were 27.1% (n=54) who us ed heroin and 47.2% (n=94), who used cocaine, 4% (n=8) who combined both hero in and cocaine, and 5.5% (n=11) who used anti-depressants/ anti-psyc hotics. Smoking was prevalen t at 72.4% (n=144) of the hypertension group, 60.3 %( n=199) were alcoho lics and over 85.4% (n=170) had history of drug abuse. Sixty point eight percent (n =121) reported had take n other prescription medications the day of thei r poisoning. Over 71.9% (n=143) of the hypertension patient were hospitalized as opposed to only 26% who were discharged form ED; .5% (n=1) died, 58.3% had major effect, and 38.7% were categorized in minor events. FIGURE 18: Prevalence of chronic hypertension compared to cardio-vascular comorbidity with poisoning. Hypertension (H TN2) and cardio vascular disease (CVD2) were similarly distribu ted in all age groups. 36
Hypertension was not correlated to th e overall poisoning exposure, however it was correlated to cocaine exposure (p= .0001). No age difference was seen that could explain this significance level. 3.3. Personal Medical History of Cardio-vascular Diseases The number of patient that presented to emergency department with a preexisting history of cardio vascular diseases (CVD) represented 17.4 % (n=113). Cardiovascular disease included all heart problems and a Pear son correlation test i ndicated heart disease was significantly correlated to cocaine abuse (r=.227, DF=645, p= .0001) 3.4. Personal Medical History of Arthritis Personal medical history of arthritis wa s only seen in 3.5% of the cases (n=23) and no significant correlation with any othe r variable was established, including the exposure to heroin that has been previously indicated to cause a degree of arthritis. 3.5. Personal Medical History of Malignancy Malignancy was rarely seen among case s of poisoning of our study; only 7 patients presented for poisoning among th e 649 cases identified, and no correlation between history of cancer and drug abuse or suicidal attempts were found. Contrary to the initial hypothesis, that patients with cancer might be at higher risk for poisoning because of multiple toxic chemotherapies, as we ll as their depressive mental state, that make them more susceptible for attempting suicide. 3.6. Personal Medical History of Asthma Asthma was of special interest to this study and it was investigated for possible correlation with multiple other variables, we identified the following findings: 37
FIGURE 19: S moking & regular drinking habits in Asthma, Depression & Psychosis Over 41.4 % of our study population had a history of confirmed diagnosis of asthma (n=269). Among the 269 asthmatics, th ere were 45.4% (n=122) who presented to ED for a complication related to heroin in take during the last 24 hours before admission. Cocaine related visits am ong asthmatics represented 30.5% (n=82), and alcohol poisoning accounted for 2.6% (n=7) of all asth ma cases. Concomitant use of heroin and cocaine represented 5.2% (n=14), and multip le drugs intoxication was seen in 4.5% (n=12) of cases. Marijuana seems to be rare as a cause for emergency visit among asthmatics and our population rate was recorded at only 1.1% (n=3.) Asthma patients who visited ED for a poisoning event had almost equal distribution of gender; 50.9% (n=135) were female, 49.1% were male. Over 72% (n=195) were single as opposed to 11.9% married (n =32). Asthma cases were predominantly African American 85.9% (n=231) with 3.3% Hispanic (n=9) and 3.7% (n=10) of white descent. Sixteen percent of our asthmatic popu lation were also diabetics, 31.6% (n=85) had hypertension, 14.5% (n=39) reported cardio -vascular problems, 3.7% of them had been diagnosed with an arthritis, and 2.6(n=7) had a past or current malignancy. 38
Depression co-morbidity among asthma patient was recorded at 18.2% and 8.6% (n=23) were diagnosed and treated for a psyc hotic disorder. Most asthma patients had a history of recreational drug use 86.1% (n =229), 24 cases (9%) were seen in the emergency department for a self poisoning. Therapeutic adverse events were also reported among asthma patients at a rate of 2.2% (n=6) and misuse of drugs at a rate of 1.1% (n=2). History of prior poisoning and ED visit or hospitalization for similar events was prevalent. In fact over 54.3% (n=146) of asthmatics were documented with prior incidents of poisoning that needed a trip to ED at least once. Inhalation was the most commonly reported mode of exposure of as thmatics 75.5% (n=203), ingestion was seen in 11.5% (n=31), then injection 6.3% (n=17) rectal or vaginal at 4.5% (n=12), and dermatological 1.1% (n=3.) Smoking was pr evalent among asthmatics at 65.1% (n=175) and regular alcohol intake was associated with asthma in 54.6% (n=147). Only 25% (n=68) of asthma patients had received ini tial management on the site of the poisoning exposure or immediately at arrival to the emergency department and 11.5% (n=31) were given immediate antid ote administration. From the total 269 patients only 14.5% (n=39) were discharged and 71.7% (n=193) were admitted to the hospital or to the ICU for further management. According to the ER physician, the assessment of the outcome reported that the poisoning events were major in 55.8% where the patient needed extra measures in order to treat their poisoning exposures, necessitating sometim es ICU hospitalization, intubation or admission to the hospital. 39
The records showed that 84% had had a follow up at the ou tpatient clinic or were referred to their primary care physician. At the time of the exposure, 67.7% of asthma patients reported that they were taking thei r asthma routine maintenance medication and 53.2% of the total asthma groups were under albute rol treatment. A significant relationship between the expos ure to heroin and history of Asthma was established (p=0.0001). Asthma was not corr elated to any of the other substances of exposure including cocaine, alc ohol, or concomitant use of heroin and cocaine. ANOVA test resulted in a statistically significant relationship between asthma and depression (p= .044) and to psychosis (.028), however smoking was not significantly associated with asthma (p= .066). TABLE-1: Summary Table for Diabetes, Asth ma and hypertension; demographics and agents responsible for poisonings. Diabetes HTN Asthma Total study population(N) 14.3% (n=93) 31% (n=199) 41.4% (n=269) Age 45.9 +/SD 11.8 47.6 +/SD 11.3 39.2 +/11.9 Gender M=60.4%, F=39.6% M=66.3%, F= 33.7% M=49.1%, F= 50.9 Heroin 29% (n=27) 27.1% (n=54) 45.4 % (n=122) Cocaine 44.1% (n=41) 47.2 % (n=94) 30.5% (n=82) Smoking 67.7% (n=63) 72.4% (n=144) 65.1 % (n=175) Alcohol 53.8% (n=50) 60.3% (n=121) 54.6% (n=147) Depression 21.5% (n=20) 17.6 % (n=35) 18.2% (n=49) Psychosis 11.8% (n=11) 11.6 % (n=23) 8.6% (n=23) Hospitalization rates 73.1% (n=68) 71.9 % (n=143) 71.7% (n=193) 40
3.7. Personal Medical History of Depression In our series Depression accounted for 21.6% (n=140) of the total cases and a significant relationship was found between poi soning and history of depression. Many patients that attempted suicide ingested a hi gher dose of their own medications for the purpose of self-harm. A smaller number presented for adverse drug reaction of antidepressor mainly Tricyclics. Over the 141 cases 27.1% (n=38) visited ED for cocaine related poisoning event and 17.1% (n=24) presen ted for heroin poisoning. The third most common poisoning seen in depressed patient s presenting to ED was anti-depressants 15.7% (n=22), multiple drugs 11.4% (n=16), salicylic acid (ASA) 9.3% (n=13), and unknown pills .7% (n=1). In patients with depressive mood disorder s, gender distribution was slightly in favor of females with 53.6% (n=75) versus 46.4% (n=65) males with a male female ratio of .8 mostly single 70.7% (n=99) versus only 11.4% of married. Af rican American was the major race in this group with 72.9% (n= 102); Hispanics were 7.1% (n=10) and 8.6% (N=12) were Whites. The majority in th e depression group were unemployed at 85.8% (n=121), 47.1% covered with IDPA and 35% were self-payers. Over 48.6% (n=68) of the depression group were brought by the EMS and 25.7% (n=36) arrived to ED unresponsive and 25% (n=35) of them had a heart related complaint, 12.9% (n=18) had a chief complain t that was classified as overdose (OD). Suicidal intent was recorded in 42.9% (n =60), and another major circumstance of poisoning in the depression group was drug a buse at 52.1% (n=73). Unintentional and misuse was only seen in less than .7% (n=1). Medication was the most commonly used in this group at 45% (n=63). The main route of poisoning in this study group of documented 41
depression patients was the i ngestion route 47.1% (n=66) a nd the inhalation route 45.7% (n=64). Rectal or injection routes we re equally recorded at 2.9% (n=4) each. Co-morbidity of depression was recorded as follow: Asthma 34% (n=48), diabetes 14.3% (n=20) hypertension 25% (n= 35) Psychosis 33.6% (n =47) cardio vascular diseases 17.9% (n=25) and arthritis 3.5% (n=5). Of the 141 patient 15% (N=21) that received immediate management at the site of the poisoning, and 8.5% received Naloxone antidote on arrival to the emerge ncy room. Beside antidote and initial intervention on the site 20% (n=28) r eceived decontamination only, 35% (N=49) received decontamination co mbined to other appropriate therapy and 36.4% (n=51) received other therapy without decontamination. Fifty-four percent of the depressi on group acknowledged smoking tobacco, 55% (n=76), regular alcohol intake, 66.7% (n=94) admitted the use of recreational drugs. Over half of the intoxication resulted in minor effects, 47.5% had so me complication and resulted in a major outcome (but not deat h), and 75.7% (n=106) were admitted to the hospital for further evaluation and management The average length of stay was at 1.4 days. Of the regular prescription medicat ion noted taken for asthma maintenance treatment, albuterol was reported by 14.2% (n =20), anti-depressant medication in 8.5 % of patients, anti-hypertensive medication 8.55 (n=12), and clonidine in 1.45 (n=2). Of the total cases 11.9% (n=77) were di agnosed and documented with a psychotic disorder. Their age varied between 20 a nd 66 years old, with a mean of 39.32 and a standard deviation of 9.55. Patients in this group were predominantly single (79.2%, n= 61) African American (83.1%, n= 64), male (58.4%, n=45) with a male to female ratio of 1.4. 42
Hispanics represented 3.9% (n=3) of cases and whites represented 9.1% (n=7); only 6.5 % (n=5) were employed. Insurance covera ge was provided by IDPA 51.9% (n=40), Medicare 7.8% (n=6), and 27.3% (n=21) were self-payers. FIGURE 20: Depression and of Psychosis (PSYCH2) by age. The age group the most involved is the middle aged between 36 and 45 years old, the same age group affected by asthma and diabetes and hypertension. 3.8. Personal Medical History of Psychosis Chicago fire department EMS brought over 56% (n=39) of them to ED and 31.2% (n=24) were walk-ins. Over 24.7% (n=19) had a heart related complaint, 13% (n=10) had chest complaint like shortness of breat h, 5.2% presented for nausea and/or vomiting. Overdose as the chief complaint was seen in 13% (n=10) and none of the psychotic patients presented for injury following an event of poisoning. Among the group of psychotics there were 14.3% of diabetics (n=11), 29.9% (n=23) hypertensive, 18.2 %( n=14) with cardio vascular diseases, 29.9% (n=23) 43
asthmatics, 62.3% (n=48) with a diagnosed depression disorder, and (1.3%) with a malignancy. Of the total patients with known psychotic disorders 42.9% (n=33) had previously visited ED or were hospi talized for a poisoning related event. Over 19.5% (n=15) of the patients with hi story of psychotic disorders had heroin poisoning, 33.8% (n=26) had cocaine poisoning,18.2% (n=18) were poisoned with their own prescription medication, and 7.8% (n=6 ) to multiple drugs combination. Alcohol resulted in intoxication in th is group only in one case (1.3%). In summary, the above poisoning exposures were distributed in 5 categories: 58.4% (n=45) illicit drug overdoses, 33.8% (n=26) pharmaceutical poisoning (though some drugs were also taken for recreationa l purposes), 7.8% (n=6) chemical industrial poisoning. Circumstances of the poisoning were collected from the History and physical (H&P) report and were as follow: suicid al 28.6% (n=22), recrea tional 64.9% (n=50) unintentional accidental 2.6% (n=2), and therapeutic e rror 2.6% (n=2). Concomitant medication intake was reported for antidepre ssant/antipsychotic for 13% (n=10), for albuterol in 11.7% of the cases, for anti hype rtension medications 9.1% (n=7) of cases and clonidine in 2.6% (n=2) of the cases. The route of exposure was recorded as fo llow: inhalation 53.2% (n=41), ingestion 35.1% (n=27), rectal or vaginal 10.4% (n=8), and injection for 1.35% (n=1). Over 15.6% (n=12) received initial management on site and 11.7% were given Narcan by the EMS technicians or immediately at arrival to ED. Decontamina tion only was ordered in 16.9% (n=13), observation only was done in 7.8% (n=6 ), 27.3% received decontamination with some other therapeutic measures, a nd 42.9 % received other therapy without decontamination. 44
Over 76% (n=59) of the psychotic patient s that presented to ED for a poisoning related complaint reported smoking tobacco, and 61% (n= 47) reported taking alcohol regularly. As many as 77.9% (n=60) of this group with documented psychotic disorders acknowledged to their ER physicians that th ey took illicit drugs. Among the total 77 patients in this group only 11.7% (n=9) of patients were discharged home, the rest (77.9%, n=60) were hospitalized. The final out come was subjectivel y assessed as major in 49.4% (n=38), as a minor effect in 46.8% (n=36); no deaths occurred in this group. The average length of stay was 1.3 days and follow up after discharge was done in 87%, through outpatient clinics and primary care physicians. 3.9. Personal Medical Histo ry of other Mental Illnesses In the process of data co llection, it was deemed logical to collect information on other mental illnesses that might be associated with depression and psychotic disorders that may influence the patients risk for self poisoning. However, the data was limited and only 3.5 % (n=19) had other concomitant psychiatric disorders such as generalized anxiety disorders, seen in 1.3% of the cases. 3.10. History of Smoking Since the review of literature has shown only rare cases of isolated tobacco intoxication where tobacco was the primary poison that necessitate d emergency visit, smoking exposure was not initially considered th e list of poisons of interest. However, an assessment of smoking status was included in order to assess its prevalence among ED related poisoning cases. Over 60.9 % (n=395) were tobacco smokers; smoking was self reported on admission and documented on the patient chart systematically as part of the initial H&P assessment. 45
FIGURE 21: Smoking and alcohol (ETOH2) co-morbidity by age. Smokers were aged between 17 and 72 years with a mean of 41.37 and a SD 10.44. Over 41.7% were female and 58.3% male w ith a sex ratio of 1.4. Of the total 395, 13.2% (n=52) were married and 71.9% (n=284) were single. Eighty two percent (n=324) of the smoking patients were African Ameri can, 5.1% (n=21) Hispanic, and 5.3% (n=21) White. Smokers rate of unemployment wa s 88.4% within our popul ation. Over 34.7% (n=137) were covered by public aid, 5.8% (n =23) by Medicare, 1.5% (n=6) by Medicaid and the rest did not have any insurance cove rage 46.1% (n=182). Patients arrived to the ED primarily by walking 29.9% (n=118) or by the Chicago Fire Department 54.2% (n=214). Their chief complaints were chest pain and hear t related complaints (33.4%, n=132), shortness of breath (25.1%, n=99), unr esponsiveness, confusion to coma (18.7%, n=74), drug overdose (11.1, n=44), vom iting (2.8%, n=11), and suicidal acknowledgement at arrival to ED for (1.3%, n=5). 46
Among the smokers group 15.9% (n=63) we re diabetics, 36.5 % (n=144) had a history of chronic hypertension, 19.7% (n=78) ha d a history of cardiovascular diseases, 3% (n=12) had chronic arthritis and there were .8% (n=3) cases of diagnosed cancer. History of depression among sm okers visiting ED for pois onings was noted in 19.2% (n=76) and history of psychosis was recorded in 14.9% (n=59). The proportion of regular alcohol drinkers among smokers was 73.4% (n =290) and a history of prior similar poisoning events that necessitated emergency visit was 47.6%. Among the same group, 77.5% (n=306) presented to ED for illicit drug related poisoning, as opposed to only 12.7% (n=50) for pharmaceutical drug poisoning and 9.1% (n=36) for industrial chemical poisoning. Tobacco smokers were poisoned with cocaine in 39.2% (n=155), heroin in 35.9% (n=142), heroin cocaine combination in 4.8% (n=19), multiple drugs in 5.1% (n=20) salicylates in 2% (n= 8) and analgesi cs in 1% (n=4). Methadone, stimulants, and marijuana all together accounted for 1.3% (n=5). The route of the poisoning exposure in smokers was mainly inhalation in 70.6% (n=179) followed by ingestion in 12.4% (n= 49), rectal and vagina l in 9.9% (n=39), and injection in 5.8% (n=23). Poison exposure in smoking patients was categorized as follows: drug and medicinal abuse for 85.1% (n=336), suicidal for 9.6% (n=38), misuse and unintentional for 1.8% (n=7) and therap eutic and adverse drug reaction for 2.5% (n=10). Over 19.5% (n=77) of cases received initial management intervention at the site of the exposure and another 13.4% (n=53) patien ts received injection of Narcan antidote. ED management among smokers consiste d of decontamination only 15.4% (n=61), observation only 6.1% (n=24), decontaminati on and other therapy in 28.6% (n=113). 47
Among this group of smoking patients, there were 46% who were managed by using other means that does not include decontamination. History of drug abuse in poisoning relate d ED visits in tobacco smokers was reported to be 91.4% (n=361). As many as 65.3% (n=258) were admitted to ICU or to a department other than ED and only 20.8% (n=8 2) were discharged; no death was reported among poisoned smokers. After discharge from the hospital 82.8% of poisoned smokers were given an appointment for follow up in outpatient clinics or with their primary physician. Concomitant consumption of other prescription medication taken for any other reason but poisoning the day of the poisoning was as follow: Albut erol 24.3% (n=96), anti-hypertensive 7.8% (n=31), anti depressant or anti ps ychotic 5.1% (n=20), clonidine in 1.5% (n=6) and analgesics 1% (n=4). Th e average length of stay among the smokers group was .9 day. 3.11. History of Alcohol Intake This study also had an interest in asse ssing the prevalence of alcohol consumption and its effect on possible intoxication ev ents, considering alcohol alone, or its involvement in potentiating other substances. Over 53.5% (n=347) reported alcohol intake routinely but the information available failed to provide a consistent a ssessment of the quantity taken and how often. Seventy four cases had unknown alcohol use and information was not available on the chart. Only 4 cases were missing altogeth er. The age ranged between 15 and 66 years, 41.5% (n=144) female and 57.1% (n=198) male with a male to female ratio of 1.3. Sixty nine point five percent (n=241) were singl e and 15% (n=52) married. Over 81% (n=282) were African American, 5.8% (n=20) were Hispanics and 5.8% (n=20) were whites. 48
Most of the poisoned alcoholic group (87% n=302) were unemployed. Only 34.9% were covered with IDPA, 44.7% (n=155) were self -payers, 6.3% (n=22) covered by Medicare, and 2% (n=7) by Medicaid. Chicago fire department (CFD) brought 51.9% (n=180), and 28.8% (n=100) of alcoholic poison related patien ts visiting ED were walk in. twenty one percent (n=73) were confused or unresponsive, 31.1% (n=108) had a chief complaint that was heart related 21.6% (n=75) had shortne ss of breath and 13.8 complained expressed an overdose related event at arrival to ED triage system. Vomiting was a chief complaint in 2.9% (n=10), suicidal attempts was found in 2.6% (n=9) of cases, and one case of injury related poisoning. Diabetes mellitu s was present in 14.4% (n=50), Chronic Hypertension in 34.6% (n=120), cardio vascular diseases in 20.2% (n=70), arthritis in 4% (n=14), asthma in 42.4% (n=147) and cancer in 1.7% (6 cases). Depression was present among alcohol users (22.2%, n=77) and diagnosed psychosis was present in 13.5% (n=47). The rate of smoking among alcoholics was 83.6% (n=290), history of poisoning events th at necessitated ED consultation was at 48.1% (n=167) and history of illicit drug us e was very high among them (92.2%, n=320). Among alcohol users, cocaine was the hi ghest poisoning exposure (39%, n=135), then heroin (33.4%, n=116); he roin. Cocaine combination was seen in 4% (n=14) of cases. Though the rate of abuse is very high in this group, it seems that alcohol using patients are cautious with their intake in that th e rate of alcohol related ED visits was very low (2.9%, n=10). It is important to note that alcohol intoxi cation might have been coded in a separate category other than poisoning, such as in an abuse category that is not part of this study. Multiple drugs were taken in 7.8% (n=27) of cases, aspirin intoxication in 2% (n=7), anti depressors and antipsychotics in 6.1% (n=21), and unknown pills, 49
marijuana, analgesics stimulants, and methadone were 2.7% (n=9) of cases. The major route of exposure reported was inhalation with 66.9 (n=232), then ingestion in 16.1% (n=56), and rectal and or vagi nal at 10.4% (n=36); in this group drug injection was seen in only 5.5% (n=19). The circumstances of poisoning was not much different from the smoking group as abuse represented 83.6% (n=290), misu se and un-intentional 2.1% (n=7), and therapeutic and adverse drug re action were at 1.5% (n=5). Suicidal circumstance was high in poisoned cases with habit of regular consumption of alcohol at 11.6% (n=41) but not as high as suicidal poisoning among smokers Twenty one point three percent (n= 74) were discharged, and 65.7% (n=228) were adm itted to the hospital. Of the hospitalized patients, follow up was offered to 80.4 % (n=279) Patient reported that other prescription medications were taken within last 24 hour s before the poisoning event as follow: albuterol in 21% (n=730), antidepressant anti psyc hotic 6.1% (n=21), and anti hypertensive in 7.5% (n=26). 3.12. History of Recreational Drug Use FIGURE 22: Prior History of illicit drug abuse by co-morbidity. Approximately 80% of the patients have had a history of prior use of illicit drug use. 50
Overall, 80% of the cases that presented to the ED reported past history or current use of illicit drugs for recreational use. Among those 80% who voluntarily acknowledged the use of recreational drugs, 39.4% (n=256) were single heroin users 35.5 % (n=231) were single cocaine users. Mariju ana was only reported in 1.7% of the cases, this very low rate of marijuana related even ts could be explained by the low toxicity of marijuana that results in very rare cases of ED poisoning related visits, however similar to alcohol, some of the marijuana events could have been coded under a different code such as abuse instead of poisoning. Twenty nine percent (n=190) of patients who reported the use of at least 2 drugs concomitantly, su ch as: cocaine concomitantly with something else 14% (n=91), heroin with something else 5.1% (n=33) alcohol with something else 5.7% (n=37) and 2.6% (n=17) who smoked mari juana while using another type of drug. 4.0. ANALYSIS OF ER MANAGEMENT & FINAL OUTCOME ED management included decontamination in only 16.9% (n=110) of cases, while combined decontamination with other therapeu tic measure was seen in 28.5 % (n=185) of cases. Seven point four per cent (n=46) benefited from observation without any other intervention. Therapeutic measures were gi ven to 43.1%. Only 2 cases were documented as refusing treatment offered (.3%). Ther e were 25 cases (3.8%) where information treatment was missing. 4.1Initial Site Management Effect on the Final Outcome Of the total patients, over 34.7% (n=128) need ed an urgent interv ention at the site of the exposure and it was documented in th e patients chart or in the Chicago Fire department transfer sheet attached to patients charts. No specific details about the intervention were collected, as it varied from simple perfusion to mechanical ventilation, 51
but the need for Naloxone administration on si te or upon the patients immediate arrival to ED was captured. Fourteen point five percent (n=92) were found unconscious and Naloxone (Narcan) administration was immediately administered at the site of the exposure or at arrival to ED in order to counteract the effect of the poison. FIGURE 23: Major Outcome and final disposition by Initial Management at the Site of the poisoning event. Both the final outcom e (OUTC_MAJOR) and the number of patients that were admitted (DISP_ADMITTED) to the hospital following a poisoning event depended on whether or not they have rece ived initial management such as Narcan injection or other treatment on the site of the poisoning. 4.2. Poisoning Final Outcome The outcome for each case is assessed and categorized into four categories: No effect; fair or minor effect; complicated or major effect; and death. Major effect was issued if the patient had been admitted to ICU, if he was injured; case resulted in permanent damage, long hospitalization or in a significant complication of his associated preexisting diseases. 52
In our series, there were 49% (n=318) major effects recorded, versus 45.3% minor effect and 3.8% were classified as no effect of the exposure on the patient. Death was recorded in .6% (n=4) cases. FIGURE 24: Rates of admissions by various co mo rbidities. Admission rates of the six above groups of patients with co morbidities ranged between 70% and 80%, but history of cardio vascular diseases was th e highest group to be hospitalized. Of the cases that were classified as major effect, there were 37.9% (n=118) female and 62.1% (n=194) male cases aged between 14 and 80 years old. Eleven point nine percent (n=38) were suicidal and 83.3% (n=265) had used an illicit drug for the purpose of recreation. Fourteen point five percent of them were married and 68.6% were singles, 80.5 % were African American. Sixt een point four percent of the total 318 patients arrived to ED unresponsive or conf used, 36.8% presented with a heart related complaint and 28.3% were short of breath. Th ere were 16.4% (n=52) were diabetics, 36.5% (n=116) had hypertension, 23% (n=73) had cardiovascular diseases and 47.2% (n=150) had chronic asthma. In this group th e adverse outcome of depression was present 53
in 20.8% (n=66), psychosis in 11.9% (n=38). Sixty seven percent (n=213) of tobacco smokers and 59.4% (n=189) of alcohol users. Heroin exposure resulted in a major outcome in 29.9% (n=95), cocaine in 44.3% (n=141), multiple drugs intoxication in 5% (n=16), and anti-depressors in 5.3% (n=17). 4.3Correlation of Outcome and Emergency Site Management A significant relationship (F= 6.035, DF=2 and p= .003) was found between the type of the initial on-site management and the final outcome. Patients who had some kind of management on the site of the exposure s eemed to have recovered better. A significant relationship was also seen between final disposition, and initial site management (F=20.37, DF=2, p= .0001). 4.4. Prior History of Poisoning Over 40% (n=265) of patients reported pr ior history of poisoning that needed a visit to ED, and were documented and verifi ed in their charts. Of these 265 cases 17.2 % (n=112) overdosed with heroin and 18 % (n=117) overdosed with cocaine. FIGURE 25: Documented prior history of poisoning and illicit recreational drug uses by comorbidity. 54
One point four percent (n=9) were poisoned with heroin in association with a second agent, and 3.7% visited the ED for cocaine poisoning concomitantly with something else. In summary, there were 6.6% (n=43) cases had a history of doubl e agent poisoning and 1.1 % (n=7) had triple agent poisoning. 4.5. Follow-Up As shown in figure 27, over 80.8% (n=525) of patients were sent for follow up to the outpatient services or to their prim ary care physician, of these 525 patients 64.8% (n=340) were sent to outpatient care and 17.5% (n=92) were sent to their primary care physician (PCP). FIGURE 26: Percent of after hospitalization follo w-up. Approximately 20% of the total cases were lost for follow up. 4.6. Final Disposition Overall, 62.9% (n=409) of patients requi red further management measures other than that received in the emergency department, and were transferred to inpatient services. Seventy eight percent were admitted to the emergency department for 55
observation, .6% (n=4) died, and the remain ing 22.5% (n=146) were discharged except for 2.2% (n=14) which were missing and may re present cases that left against medical advice or eloped from the emergency department. More males 62.1% (n=90) than females 37.9% (n=55) were discharged from ED, and the age median was 40 years for this group of patients. This may indicate that more females were hospitalized because their poisoni ng event looked more serious or because they did not respond to the initial management in ED. Figure 27 : Final disposition distribution. Most patients were transferred to another department in the same hospital, approximately 10% were admitted to ED for observation and around 20% were discharged. Among the group of patients that were di scharged from ED, 79.5% (n=116) of them had an event related to the use of a recreational drug, 1 5.1% (n=22) from the use of pharmaceutical drugs and 4.8% (n=7) were poiso ned with a chemical that was industrybased. 56
In the Admitted group, patients ages varied between 1 and 85 years, mean 40.69 (+/SD 13.067), 45.2% were females and 54.8% were males; 70.2% of them were single and 11.2% married; 65% were exposed to il licit drugs and 23.5% to pharmaceutical drugs. Over 47.2% (n=193) were asthmatics, 35% (n=143) had chronic hypertension, 22.7% (n=93) had cardio vascular disease, 4.4% (n=18) had chroni c arthritis, and 1.2% (n=5) had a malignancy. Depression was pr esent in 25.9% of the hospitalized group, psychosis in 14.7% (n=60), smoking in 63.1% (n=258), and alcohol use in 55.7% (n=228). Suicidal poisoning were present in 16.1% (n=66) and abuse in 73.8% (n=302); their average length of stay was 1 day. FIGURE 28: Admissions and discharge were highest during the first shift then decline sharply during the third shift (Disp_Disch: disposition dischar ged; Disp_Admitted: Disposition Admitted to hospital, disposit. 4= died). 57
58 5.0ANALYSIS OF CORRELATION One of the primary interests of this st udy was to explore the significance of the relationship between asthma and heroin use in the population of ED that was reported in the literature, for which many mechanisms were speculated, but none confirmed. In fact Chicago has one of the highest mortality rates and prevalence of asthma, and also one of the highest prevalence of drug abuse which makes these two variables of special interest for this study. After checking the observations independe nce and verifying the normality of the population, bivariate analysis wi th computation of both a Pear son test and a (2-tailed) t test was performed. It is a measure of correla tion that varies between -1 and +1, with 0 indicating no relationship (random pairi ng of values) and 1 indicating perfect relationship, i.e. "The more the x, the more the y, and vice versa." A value of -1 is a perfect negative relationship, i.e. "The mo re the x, the less the y, and vice versa. R square is the coefficient of determination, the square of the Pearson correlation coefficient, and it represents the percent of the variance in the dependent variable explained by the independent. The significance level of the Pearson test was set at 0.01 After bivariate analysis, a partial corr elation test was performed. Both the bivariate and the partial test have shown a statistically si gnificant relationship between asthma and heroin use with coefficient at (r= .175, DF=647, p=0.001). This relationship is positively correlated, indicati ng that an increase of one va riable results significantly in an increase of the second and may be explained as such: the overuse of heroin may exacerbate asthma. Or the opposite effect coul d also be true: when having a history of asthma this may predispose this group of patients to over-use heroin, knowing that the
latest research is pointing toward genetic predisposition for both of them (Matthias 2007, Xuei et al., 2007, Balaci, 2007 ) The bivariate correlation (2 tailed) t test showed a strongly si gnificant correlation between heroin and asthma w ith (r=.175, p= .0001). An r partial Pearson correlation test, which gives the advantage of correcting for the effect of other confounding factors by blocking the effect of other va riables that may participate in the correlation of interest, was also performed. After correcting for the effect of de pression, psychosis, smoking, alcohol, ASA exposure, prior history of poisoning, anti-depre ssor and anti-psychotic drugs, methadone, and cocaine exposure, it wa s found that the significance between asthma and heroin disappeared, and the Pearson coefficient dr opped to .074, the p valu e increased to 0.061, becoming statistically non significant. This coul d be interpreted to mean that there was no true correlation between va riables of interest. 5.1Depression Correlations Depression was significantly correlated w ith the variable su icidal attempt (r=.275, DF=647, p=.0001), and with multiple drugs exposu re, (which is most likely deliberate intake for the purpose of suicide or abuse), (r=.136, DF=645, p=.001). Another significant relationship was found between depr ession and history of psychosis (r= .352, DF=647, p= .0001) and between depre ssion and the exposure to antidepressors/antipsychotic drugs (r= .215, DF645, p= .0001), as well as with the use of ASA (p=.172, DF=645, p=.0001). Heroin expos ure was negatively correlated to depression (r=-.200, DF=645, p= .0001). 59
As far as the route of exposure the biva riate analysis has shown that ingestion route was significantly correlated (r=335, DF=647, p=.0001), inhalation route was also found significant but it was negativel y correlated (r=-175, DF=647, p=.0001). Disposition was categorized in discharge, admission to the hospital and death. The analysis revealed that depression and di scharge from the hospital were negatively correlated (r=-.103, DF=647, p= .009). Depression and admission to the hospital were positively correlated (r= .138, DF=647, p=.0001), which simply translates to an unfavorable prognosis of poisoning when depres sion was present. This could also be due to the higher dose intake of the drug due to th e presence of depression or it may be due to the choice of the drug itself that is differ ent when the patient is depressed and having suicidal thoughts. The poisoning agents were categorized as: pharmaceutical drugs/medications, chemical domestic, chemical industrial, or occupational and recreational exposure. Only pharmaceutical drugs/medications were signif icantly correlated to depression (r= .307, DF=647, p=.0001). Suicidal poisoning were positively correlated (r= .453, DF=647, p=.0001), abuse was also significantly correlated to depression (r= -.305, DF=647, p=.0001) TABLE 2 : Summary list of depressi on significant Correlations. Variable 1 Variable 2 R DF p Value Depression ASA .172 645 .0001 Heroin -.200 645 .0001 Admission to Hospital .138 647 .0001 Pharmaceutical drugs .307 647 .0001 Suicidal poisoning .453 647 .0001 60
5.2Psychotic Disorders Psychosis was significantly co rrelated to depression as previously mentioned, to heroin (r= -121, DF=645, p= .002), to sm oking (r=.118, DF=647, p= .003), to insurance coverage (particularly IDP) (r=129, DF= 647, p= .001) and self-coverage (r= -.137, DF=647, p=.0001). Similar to depression psychot ic disorders were also found to be significantly correlated to pres cription medication of anti-de pressant and anti-psychotics (r= .190, DF=645, p=0001), which is generally their prescription medication and to ingestion the route (r=.125, DF=647, p= .001). Admission disposition is al so strongly correlated with psychotic disorders (r= .113, DF=647, p= .004) as well as to pharm aceutical drugs (r= .114, DF=647, p= .004). Suicidal circumstance was also significantly co rrelated with psychot ic disorders (r= .163, DF=647, p= .0001) and to abuse circumstances (r= .102, DF=647, p= .009). TABLE 3 : Psychotic disorders significant correlations with poisoning. Variable 1 Variable 2 R DF p Value Psychosis Co-morbidity Smoking .118 647 .003 Insurance IDP .129 647 .001 Anti-Depr/Anti-Psych .190 645 .0001 Ingestion route .125 647 .001 Admission .113 647 .004 Prescription drugs .114 647 .004 Suicidal poisoning .163 647 .0001 Illicit drug abuse .102 647 .009 5.3Heroin Poisoning Correlations Heroin was found to be correlated to ma ny other variables in our population, such as depression, psychosis, cardio vascular diseases (r= -.181, DF=644, p= .0001), to cocaine (r= -.508, DF=645, p= .0001), to abuse of heroin and cocaine combination (r= 61
.158, DF=645, p= .0001), to multiple drugs exposure (r= -.180, DF=645, p=000), history of drug abuse (r=286, DF=635, P=000), to the Hispanic race (r=-146, DF=645, p=000), to self insurance (r=212, DF=644,p=000) to arrival by CFD (r=213, DF=643, p=000), arrival to ED by walking (r= -.143, DF= 645, p=000), chief complaint Unresponsive (r=152, DF=645, p=000), chief complaint heart related (r=0.268, DF=645, p=000), shortness of breath chief complaint (r= 177, DF=645,p=000), Aspirin intoxication (r= .139, DF=645, p=000), anti depressor/anti-psychotic exposure (r= -139, DF=645, p=000), ingestion route (r=-359, DF645, p=000), rect al vaginal route (r=.350, DF= 645, p=000), management at the site of the exposure (r =.111, DF=630, p=000), suicidal poisoning (r= .234, DF=245, p=000), circumstance abuse (r = 324, DF645, p=000) and concomitant intake of other substances (r=.258, DF=645, p=000). TABLE 4 : Summary of heroin poisoning correlations. Variable 1 Variable 2 R DF p Value Heroin Poisoning Vaginal or rectal route .350 645 .0001 History of drug abuse .286 635 .0001 Heart related CC .268 645 .0001 History Illicit drug abuse .324 645 .0001 SOB (CC) Asthma .117 .174 645 645 .0001 .0001 Depression -.118 644 .0001 Suicidal poisoning -.234 645 .0001 5.4Cocaine Poisoning Correlations Cocaine was found to be significantly correlated to chronic hypertension (r= .220, DF=745, p=.0001), and to cardiovascular diseases (r= .227, DF=645, p=0001. Cocaine was also significantly correlated to heroi n, to heroin cocaine combination, (r= -145, DF=745, p=000), to multiple drugs (r=-.166, DF=645, p=000), to smoking (r=201, 62
DF=645, p=000), and to the habit regular al cohol intake (r=.165, Df=645, p=000), to history of drug abuse(r= .245, DF=645, p=000, ), to age (r= .227, DF=645, p=000), to race White (r=-.140, DF=645, p=000), arri val mode through CFD (r=-149, DF=645, p=000), arrival walking (r=173, Df=645, p=000), chief complaint unresponsive (r=-.234, Df=645, p=000), chief complaint heart related symptoms such as chest pain (r= .507, Df=645, P=000), chief complaint vomiting (r= -.200, Df=645, p=000), anti depressors and anti psychotic exposure (r=-.173, Df=645, p= 000), ingestion route (r= -.353, DF=645, p=000), inhalation (r=.430, Df=645, p=000) to rectal/vaginal (r=-.225, DF=645, p=000), to outcome minor (r= -.222, DF=645, p=000) to major outcome (r= .266, Df=645, p=000), to discharge disposition ( r= -.274, DF=645, p=000), admitted disposition (r=.181, DF=645, p=000), exposure to medicat ion and pharmaceutical products (r=-.353, DF=645, p=000), to recreational drugs (r=283, DF=645, p=000), suicidal category of exposure (r= -.258, DF=645, p=000). TABLE 5 : Significant correlations with cocaine poisoning. Variable 1 Variable 2 R DF p Value Cocaine poisoning Inhalation .430 645 .0001 Major outcome .266 645 .0001 Chest pain .507 645 .0001 History of drug abuse .245 645 .0001 Age .227 645 .0001 History of Illicit drug abuse .283 645 .0001 Unresponsiveness -.234 645 .0001 Hospital admission .181 645 .0001 Antidep/antipsych .173 645 .0001 Prescription medications -.353 645 .0001 63
6.0. PREDICTION ANALYSIS One of the interests in this study is to determine predic tors for the final outcome of ED poisoning, which could be useful for improving patients management and also in implementing preventive measures. Predicto rs for poisoning final outcome, asthma, and for the risk of suicide in our population we re computed using re gression analysis. The stepwise regression analysis for outcome resulted in the construction of the following models. Final Outcome of Poisoning = 0.108 + 0.200 Cocaine + 0.114 Inhalation + 0.121 CVD + 0.103 Asthma + 0.94 Gender + 0. 09 Prior Poisoning History Cocaine was the most predictive for the major outcome of poisoning, route of inhalation was the second most important pred ictor. Cardiovascular diseases were third, asthma, gender and prior history of poisoning. Stepwise regression analysis was also run to compute predictors for the risk of suicide attempts using poison. Predictors in tercept and coefficient, as well as the significance of the overall model and the in dividual models produced, as well as the regression assumptions described in addendum 1. Measures of association are expressed in the following mathematical formula: Model (Suicide) = 0.394 +0.330 *Depressi on -0.005* Age -0.99* Gender+ 0.072* Prior poisoning history-0.055-0.082*unemployed Depression is the greatest ri sk factor for suicide attempts; age and gender, prior poisoning event, and employment were also significantly correlated to suicide, and 64
remarkably, psychotic disorders, though know n for their possible risk for suicide, particularly, in younger schizophrenics, our data did not support this assumption. A stepwise regression was performed for bot h asthma as a dependant variable and heroin as an independent with the assumpti on that heroin affects asthma and exacerbates it. The measure of association was expresse d in the following mathematical formula: Model (Asthma) = 0/419 + 0.221 *Inhalation + 0.147 Prior Poisoning History. -0.140 Gender + 0.149 Heroin + 0.728 Cancer + 0.136 Race_AA 0.004 Age Again, inhalation is the greatest predictor of asthma with prior history of abuse, heroin, gender, age, and race of African American being the most other important predictors for asthma (Table 9, Table 10). Similarly, another regression was run and asthma was the independent variable to check on the assumption that asthma predispos es patients to abusing heroin. The measure of association is expressed in the following formula: Model (Heroin) = 0.284 + 0.794* Rectal/Vaginal route + 0.142* Asthma + 0.273 Inhalation + 0.394*Injection 0.193* CVD 0.110* Prior documented Poisoning history 0.077* Alcohol + 0.111 Psychosis Asthma was second strongest heroin abuse predictor (Table 3, Figure 41, 42, 43). This could also be that the observed correla tion might indicate asthma is predisposing heroin addiction, or at least th at the overall associ ation is just a mixed effect from both variables in both directions Inhalation is also a pred ictor of heroin abuse. Rectal and vaginal route of drug use is also very prevalent in heroin abusers. Psychosis is also a predictor of heroin abuse, wh ich supports the study s third hypothesis. 65
DISCUSSION Poisoning in the emergency department is still a problem, especially when it relates to the use of recreational drugs. The magnitude of th is problem has been explained elsewhere, however its implications in ED have been actively illustrated in this investigation. In exploring the relationship between asth ma and heroin, our data revealed that there are two other significantly correlated co-morbidities wi th drug abuse, particularly chronic hypertension and cardiovascular diseases that were correlated to cocaine use. In reviewing the literature, we were unable to find any evidence of similar correlations, though, in our study it was found to be st rongly correlated (r= .220, DF=745, p=.0001) for hypertension and (r= .227, DF=645, p=0001) fo r cardio vascular diseases. While the relationship of asthma to drug abuse is complicated, and many authors have suggested numerous theories to explain how heroin corr elates to asthma, hypertension correlation to cocaine might be due to a more direct effect of cocaine on vessels. In fact, cocaine was extensively studied, especially its effect on enzymatic modulati on and its ability to affect the fate of cell necrosis and apoptosis in the liver ( Harbison et al., 2003, Harbison et al. 2000, Price et al. 1999 ). Cocaine has also been shown to have numerous other effects on the cell signaling system and on the vascular bed by modifying the pe ripheral resistance and inducing a potent vasospasm that raises the blood pressure and increases the cardiac output. The vasopressor effects of cocaine are mostly mediated by norepinephrine of sympathetic nervous system (Hollander et al 2008). By the same mechanism, cocaine 66
induces a vasoconstriction in coronary vessels and results in chest pain that mimics unerringly an angina or a myocardial infarc t. Cocaine abusers may present chest pain similar to angina or corona ry infarct which could result in a hospitalization. On discharge, a patient will gene rally be categorized as having a history of chest pain that will be documented in the patients medical history but without any mention of its temporal relationship to cocaine This type of documentation might have interfered with a more accurate heart assessment and resulted in significant misclassification of cases in the cardiovascular diseases gr oup. As a result, an apparent co rrelation of cocaine to heart disease might have been seen as a result of this misclassification. In summary, the apparent correlation of hyperten sion to cocaine might just be due to the direct effect of cocaine on the vasculature and in the case of cardiovascular disease and it seems that some of these cardio vascular co morbiditie s that were captured were partially, if not totally, cocaine-related diagnoses In a study conducted in ED to assess the prevalence of cocaine use in patients with hypertension who presented to ED for co mplaints not related to cocaine use, the author reported that only 13% of the tota l 99 cases of hypertension tested positive for cocaine, and 38% of them presented for chest pain or for congestive heart failure and were prescribed beta-adrenergic antagonist me dications. In our series, the prevalence of cocaine admissions among hype rtension patients was 47.2% (n=94) which is higher compared to what was previously reported. As a result of the high prevalence rate of heroin and cocaine use in our population, it is unlikely that patients w ho have underreported their drug use have affected the results of our study. The higher prevalence of drug use cannot be explained 67
by over reporting because patients are more likely to underreport th an to over-report the use of controlled substances. However, even if they have unde rreported, one would expect our observed difference to be even la rger. Thus, if any drug controlled user was incorrectly misclassified into a non-user group rather than a user group, our results would be expected to show a higher difference. Many cases were reported in recent years in relation to sudden and severe asthma incidents that were temporally correlated to the abuse of co caine and heroin, and numerous authors report this relationship. They further c onclude that heroin may be triggering asthma attacks and that heroin a ssociated withy status asthmaticus. Another study reviewed death reports in the state of Maryland and demonstrated that a high percentage of patients who died from asth ma had also abused drugs and their post mortem urine drug tests were positive (Weitzman et al.). Over 4% of all admissions for asthma in the US require mechanical ventilation, though there is still controversy regarding the optimal treatment for severe asthma requiring intensive care optimal intubations; some authors ha ve also reported that many of the asthma patients requiring intubations are heroin abusers. Asthma patients, it seems, tend to use various drugs for abuse, and both legal and illicit drugs were described to be abused by asthma patients, including illegal inhalers (Thompson et al., 1983) and prescription drugs (including the ones that were not widely used by others such as glucorti coids, and inhalers salbutamol) (Kinirons, 1983. Quddusi et al. 1998). There was also a study conducted that was design ed to compare tobacco and alcohol use in asthmatic students to non-as thmatics. This study found that high school students with current asthma used cigarettes, cigars, marijuana, and inhalants (huffing) at 68
rates greater than high school students without current asthma (Boyd et al. 2006; Jones et al. 2006). It appears that asthma patients ar e subject to higher abuse susceptibility; this might be due to many reasons. The high anxiet y of asthmatics could drive asthma patient to reduce their level of anxiety by abusing various drugs There is still much to know about asthma co-morbidities, in particular its associated behavioral changes and effect on patients psychological well being. Recently Blackman et al. (2007) reported that children with asthma have higher rates of attention deficit/hype ractivity disorders; diagnoses of depression, behavi oral disorders, learning di sabilities, and missed school days. Behavioral changes were also seen in the adult population, and the effect on asthma management has always been a concern for in creased morbidity and mortality in asthma. (Volicer et al. 2006). Moreover, there is evidence th at drug dependence can run in families and that a person's genetic makeup can affect how the body processes a drug, and could also determine the drug's effect. These individual differences can make some people more likely to use drugs and to become dependent upon them. Due to its prospect in therapeutic and its role as an analgesic, heroins mode of action and mechanism of abuse and dependence has been heavily studied; particularly its genetic aspect where basic animal research and human studies have shown that complex genetic mediators exist, many gene and receptors are implicated in this behavioral aspect of heroin addiction. (Xuei et al., 2007) The relationship between asthma and heroin poisoning has been seen in previous reports from an asthma management point of view; heroin and asthma were observed to be temporally related. It was then concluded that heroin exac erbates asthma but a mechanism could not be esta blished. Though this approach seems simple but logical, 69
researchers are skeptical of a receptor mediat ed effect on the bronchus as opiates favor blocking bronchoconstriction through these receptors instead of stimulating it. Researchers have also argued that the heroin effect on bronchocons triction might occur through a non-receptor mediated mechanis m that might also potentiated by an inflammatory response. No supporting evidence has been produced regarding this theory other than heroin may deregulate ma st cells and rele ase histamine. Depression has been described in the liter ature as a risk factor for poisoning because of the suicidal tendency that genera lly accompanies depressed patients. Our data has reinforced this belief and illustrated a strong relationship between suicidal intent using poison. Suicidal intent was found to be positively correlated (r=.453), which statistically translate into an r squared that is equal to .205. This value may be described as: over 20% of suicide events seen in ED were attributed to this preexisting depression. Though depression as a risk factor has been pr eviously reported, little description of the magnitude of the depression contributes in causing patients to commit suicide using poison. In our series, and according to our data, this component was estimated to 20.5%. It is known that there are certain triggers, for depressed patients with suicidal thoughts, which intensify suicidal urges. Th e question that remain s unanswered is how much drug addiction is related to the modification, in tensification, or renewal of suicidal triggers in comparison to non-addicted, depressed or ps ychotic patients. How much impact does drug addiction have on decisi on capabilities regarding self harming, assuming that he or she is in fact under ex tremely high stress from his or her need for drugs, and additional stress that originates from his or her depressed mood or psychotic 70
illusion. If that is the case, what determines the patients behavior in choosing the dose, the method, and the timing for shifting from intent to action? In this study, psychotic disorders were also identified as being significantly correlated with suicide attempts, but the coefficient of significance was relatively smaller than the one seen in depression. The Pearsons coefficient was r =.163 which translates to only 2.6% of suicidal effect th at can be attributable to pr eexisting psychotic disorders. However, another correlation was also established between psychosis and the use of antipsychotic and anti-depressors that could indicate that patie nts with psychotic disorders are subject to the risk of auto intoxicati ng themselves by using their own prescription medication. A similar pattern was seen in pati ents with depression that are under tricyclic medication, but this scenario is even more complicated with anti-depressors given the fact that these medications have a very narrow safety window, and many depressed patients overdose unintentionally. Patient educ ation regarding tricyclic toxicity has become routine now after all th e accidents that were seen in ED. Moreover, the current recommended best practice is to limit the pr escription of tricyclics antidepressants in favor of the newer selective serot onin reuptake inhibitors (SSRI). Poisoning and drug addiction remain a significant burden on the healthcare system, particularly on the asthmatic. Our fo llow up cost study was designed to compare asthmatics with a history of drug addiction and poisoning to anothe r group of asthmatics without any history of drug poi soning or addiction. Both groups were compared with respect to length of stay (L OS) inpatient, age, gender, and expenditures. The mean LOS for asthmatics without drug abuse was 1.21 days and a SD of .695, and for the asthmatic patient with history of illicit drug poisoning of 7.07 days. The hospital charges ranged 71
between $99.00 and $18096.00 with a mean of $2941 for asthmatic without abuse and between $52.00 and $695874.00 with a mean of $28028.00 for the group of asthmatics with illicit drug poisoning. A significant difference was established (p<.001) between the LOS of asthma with heroin and/or cocaine ab use as compared to LOS of asthma without abuse, and a significant difference in hospita l cost was also established between the 2 groups (p< .0001.). There are several potential limitations to this study. The study is retrospective and the information was collected from the patien ts charts, which varied in accuracy and availability between cases. Due to the high ra tes of drug abuse hist ory, it does not seem that this study underestimated the abuse; how ever it might have underestimated some comorbidity such as arthritis, and particularly depression, which is still not screened systematically for every patient visiting ED especially in a se tting of drug overdose emergencies. Another potentia l limitation of this study was that there was no way for the study to follow up on specific cases and screen patients for suicidal ideation to find out if they have had any intent for self-harm. One of the other limitations to this study is inherent to case capture, given the initial design that did not take in consider ation the complexity of the coding process. Unfortunately this issue was discovered after th e data collection starte d; in fact cases like alcohol poisoning were not coded under poisoning, but under abuse, and similarly, marijuana use was not captured under poisoning, but under a separate co de. In the case of alcohol and marijuana use, our study has accoun ted mainly for events with overdose that resulted in significant effects. Similarly, adverse drug reaction and therapeutic poisoning were not coded under poisoning unless they resu lted in a significant effect. The study aim 72
was not adverse reaction, marijuana or alc ohol abuse, which was the reason why these deficiencies did not affect the validity of the st udy results nor its abil ity to be generalized. 73
CONCLUSION In conclusion, the findings of this res earch supported the hypothesis that drug abuse is the greatest risk factor for po isoning in the emergency department over prescription drugs, environmental and occupati onal exposures. This research has shown that there is correlation between chronic dis eases and intentional poisoning using illicit drugs, in particular cocaine and heroin. This study has al so shown a correlation between chronic hypertension and cocaine cardiovascular diseases an d cocaine, and a correlation between asthma and heroin abuse. Our data ha s verified that there is an ongoing change in illicit drug route of intake that is shifting from injection to inhalation; expectantly this shift will have a positive effect on reducing th e prevalence of HIV and hepatitis B in the future. However, it seems that the heavy use of illicit drugs through inhalation has resulted in novel problems to asthma patien ts. A measure of associ ation was established between asthma and heroin use, and inhalation was found to be the greatest predictor for asthma in this population. The results suggest that suicidal behavi ors are not infrequent occurrences among drug users, and that they often coexis t and complicate other medical conditions. Predictors for suicide attempts were produced, and it was determined that interventions designed to reduce suicidal be haviors should simultaneously address preexisting medical conditions that could predispose or increase the patients risk for transitioning from ideation to action. Rates of depression and a ssociated psychotic diso rders in drug addicts 74
with asthma indicates a need to systemati cally screen young and middle aged adults with asthma and drug abuse for depression and for suicide risk. This study provides evidence supporting the idea that heroin and cocaine abuse are serious issues that significantly raise the cost of asthma and increase both the rate of hospitalization and hospital LOS. Pragmatic guidelines and innovations in reducing heroin and cocaine abuse in patients with asth ma may improve the severity of the disease, and reduce its burden on the hea lthcare system and on society. This research was able to characterize th e circumstances of incidents of poisoning and provide a greater understanding of the extent of th is problem and its burden on patients and on society. It has also identifie d numerous predictors that can be used by healthcare organizations, practitioners, a nd health policy makers for improving the management of asthma and its complications that result from drug use. This study has helped in consolidating a c onsiderable amount of information on co-morbidity, drug use, smoking and alcohol prevalence that may be very useful for patient management and future research. Drug abuse is a burden on asthma patients, particularly those that are not covered by insurance. Asthma education and se lf-management techniques in patients acknowledging the use of recreational drugs specifically heroin or cocaine by insufflations, need to be provided and patients must be informed about the potential risk and educated on how to prevent significant asthma exacerbation by avoiding common triggers, drug use, and other additional stress. Enrollment of asthma patients in free drug rehabilitation programs may be the way of the future; these progr ams can be sponsored by the state or by insurance companies. Sim ilar programs will cut down significantly on 75
asthma costs that are still a top public health concern in many states such as the state of Illinois. The magnitude of loss from the additio nal burden that poisoning is placing on the society and on insurance companies, as well as on asthmatics, can be minimized if similar steps were taken. 76
REFERENCES Balaci Lenuta, Maria Cristina Spada, N azario Olla, Gabriella Sole, Laura Loddo, Francesca Anedda, Silvia Naitza, Ma ria Antonietta Zuncheddu, Andrea Maschio, Daniele Altea, Manuela Uda, Sabrina Pilia, Serena Sanna, Marco Masala, Laura Crisponi, Matilde Fattori Marcell, Devoto, Silvia Doratiotto. (2007). IRAK-M Is Involved in the Pat hogenesis of Early-Onset Persistent Asthma. The American Journal of Human Genetics; 80:1103-1114. Bernstein J, Bernstein E, Tassi opoulos K, et al. (2005) Brief motivational intervention at a clinic visit reduces cocaine and heroin use. Drug Alcohol Depend; 77:49-59. Bialas MC, Reid PG, Beck P, et al. Changi ng patterns of self-poisoning in a UK health district. QJM 1996; 89(12):893 901. Blackman JA, Gurka MJ. (2007). Developpement al and behavioral comorbiditeis of asthma in children. J Dev Behav Pediatr; 28(2):92-9 Bouknight Reynard, MD, PhD, Patrick AlGuire, MD, Richard Lofgren, MD & Ruth Hoppe, MD. (1985). A profile of the self poisoner in Michigan. AJPH Dec 75 (12) 1435-36 Boyd J.J, M.J. Kuisma, A. O. Alaspaa, E. Vuori, J. V. Repo & T.T. Randell. (2006). Recurrent opioid toxicity after pre-hosp ital care of presum ed heroin overdose patients. Acta Anesthesiologica S candinavica. 50(10), 1966-1270 77
Boyd, carol J., McCabe Sean Esteban, Teter Ch ristian J., (2006). Asthma Inhaler misuse and substance abuse: a random surv ey of secondary school students. Addictive Behaviors; 31(2), 278-287 Byard RW, Klitte A, Gilbert JD. Changing patt erns of female suicide: 1986 2000. J Clin Forensic Med 2004; 11(3):123 128. Center for Disease control and preventi on (CDC), unintentional and undetermined poisoning deaths-11 states, 1990-20 01.MMWR Morb. Mortal Wkly. Rep. 2004 53(14)-238 Centers for Disease Control and Prevention (CDC). Unintentional and undetermined poisoning deaths-11 states, 1990 2 001. MMWR Morb Mortal Wkly Rep 2004; 53(11):233 238. Cook Magdalena, MS, Paul A. Simon, MD, MPH, & Richard E. Hoffman MD, MPH. (1995). Unintentional carbon monoxide poisoning in Colorado 1986 through 1991. American journal of Public Health ; 85(7) 987-990. Cygan James, MD, Matthew Trunsky MD, & Thomas Corbrige, MD, FCCP. (2000). Inhaled heroin-induced status asthma ticus. Chest; 117:272-275. D'Onofrio Gail, MD, MS, Bruce Becker, MD, Robert H. Woolard, MD, (2006). The Impact of Alcohol, Tobacco, and Other Drug Use and Abuse in the Emergency Department. Emerg Med Clin N Am 24:925-967 Dawn N Albertson, Carl J Schmidt, Gre gory Kapatos, and Michael J Bannon. (2006). Distinctive Profiles of Gene Expre ssion in the Human Nucleus Accumbens Associated with Cocaine and Heroin Abuse; Neuropsychopharmacology; 31(10): 2304-2312. 78
Gahlinger P. Club drugs: MDMA, ga mma-hydorxybutyrate (GHB), rohypnol, and ketamine. Am Fam Physician 2004;69:2619-26. Gulash Seydaoglu, MD, Salim Satar, MD, & Nazan Alparslan, Ph.D. (2005) Frequency and mortality risk factors of acute ad ult poisoning in Adana, Turkey, (19972002). The Mount SinaiJournal of Medi cine; 72(6):393-401 Gulsah Seydaoglu, M.D., Salim Datar, MD, Nazan Alparslan, P.h.D, frequency and Mortality Risk Factors of Acute a dult Poisoning in Adana, Turkey, 19972002. Mount Sinai Journal of Medicine 2005, Vol.72, 393-401. Gupta heeraj, Brian Keogh, Kian Fan Chung, Jon G. Ayres, Caroline Goldfrad, Anthony R Brady, & Kathy Rowan (2004). Characterestics and outcomes for admissions to adult, general critical care unit with acute severe asthma. Crit Care. 8(2):R112-R121 Harbison, R. D., Babu, S., Muro-Cacho, C. A., Francois, M. R., and Stediford, T. (2000). Adrenergic Modulation of Brom ohydroquinone-Induced Nephrotoxicity. Research Communications in Pharm acology and Toxicology 5. Harbison, R. D., Stedeford, T., Muro-Cacho, C., Mosquera, D. I., and Banasik, M. (2003). Activation of alpha1-adren ergic receptors potentates the nephrotoxicity of ethylene dibromide. Toxico logy 186, 181-189. Hawton K, Fagg J. (1992) Trends in deliberat e self-poisoning and selfinjury in Oxford, 1976 1990. Br Med J; 304:1409 1411. Hawton, Keith, Sutton, Lesley, Haw, Camilla, Si nclair, Julia, Harriss, Louise. (2005). Suicide and Attempted Suicide in Bipolar Disorder. Journal of Clinical Psychiatry 66(6):693-704 79
Hawton, Keith, Sutton, Lesley, Haw, Camilla, Si nclair, Julia, Harriss, Louise. (2005). Suicide and Attempted Suicide in Bipolar Disorder. Journal of Clinical Psychiatry, 66(6): 693-704 Henry, John A., Guy's & St Thomas' Suicide risk and antidepressant treatment. (1996). Journal of Psychopharmacology, 10(1): 39-40 Henry, John A., Guy's & St Thomas' Suicide risk and antidepressant treatment. (1996). Journal of Psychopharmacology, 10(1): 39-40 Hollander Judd E. MD, (2008). Cocaine in toxication and hypertensions. Ann Emerg Med; 51(3 Suppl): S18-20 Hughes PH, Storr CL, Brandenburg NA, et al. (1999). Physician substance use by medical specialty. J Addict Dis;18:23-37. Ichimura, Atsushi, Ichimura, Atsushi, Matsumoto, Hideo Aoki, Takayuki, Andoh, Hidehiro, Yano, Hiroshi, Nakagawa, Yoshihide, Yamamoto, Isotoshi, Inokuchi, Sadaki. (2005). Characteristics of Suicide Attempters with Depressive Disorders. Psychiatry & Clinical Neuroscienc Jason WW. Thomason, Ayumi Shintani, Josh f Peterson, Brenda C Jackson & E Wesley Rly 92005). Intensive care unit delirium is an independent predictor of longer hospital stay: a prospectiv e analysis of 261 non ventilated patients. Crit Care; 9(4): R375-R381. Jones SE, Merkle S, Wheeler L, Mannino DM, Crossett L. (2006). Tobacco and other drug use among high school students with asthma. J Adolesc Health; 39(2):291-4. 80
Kingsbury, S: Do adolescents who take ove rdoses have specific psychological characteristics? A comparative stud y with psychiatric and community controls. (1999). Journal of the American Academy of Child and Adolescent Psychiatry, 38(9):1125-31 KINIRONS M J, (1983). Addicti on to aerosol treatment: the asthmatic alternative to glue sniffing British Medical Journal; (287): 1515-1516. Koplan Jeffrey P., M.D., David W. Flemi ng, M.D., Barbara R. Holloway, M.P.H., John W. Ward, M.D., Caran R. Wilbanks (Heroin Overdose Deaths Multnomah County, Oregon, 1993-1999 MMWR, Morbidity and Mortality weekly report, CDC, July 21, 2000/vol.49/No 28) Krantz Anne J. M.D., Ronald C. Hershow, M.D., Nikhil Prachand, M.s., Donna M. Hyden, M.P.H, Cory Franklin, M.D., Daniel O. Hryhorczuk, M.D, Heroin Insufflation as a trigger for patients with Life treatning Asthma. CHEST 2003; 123: 510-517. Krantz Anne J., MD, MPH, Ronald C. Hershow, MD, MPH, Nikhil Parachand, MS, Dona M. Hayden, MPH, Cory Franklin, MD & Daniel O. Hryhorczuk, MD, MPH. (2003). Heroin insufflation as a tr igger for patients with life-threatening asthma. Chest, 123:510-517. Lachheb J., Difallah I.B, Chelbi H., Ham zaoui K. & Hamzaoui A. (2008). Toll-like receptors and CD14 genes polymorphisms and susceptibility to asthma in Tunisian children; Tissue Antigens (1) doi:10.111/j.13990039.2008.01011.x.). 81
Levine Michael, M.D., Maria Elena Iliescu, M.D., Helena Margellos-Anast, M.P.H., David A. Ansell, M.d., (2005) The eff ect of cocaine and heroin use on intubation rates and hospital utilization in patients with acute asthma exacerbations. CHEST. 128:1951-1957 Levine Michael, M.D., Maria Elena Iliescu, M.D., Helena Margellos-Anast, M.P.H., David A. Ansell, M.d., The Effect of Cocaine and Heroin Use on Intubation Rates and Hospital Utilization in Patients With Acute Asthma Exacerbations. CHEST, 2005; 128:1951-1957 Litovitz TL, Felberg L, White S., Klei n-Schwartz W: 1995 Annual Report of the American Association of poison contro l centers toxic exposure surveillance system. Am. J. Emerg. Med. 1996; 14: 487-537. Litovitz TL, Klein-Schwartz W, White S, et al. 2000 annual report of the American Association of Poison Control Centers toxic exposure surveillance system. Am J Emerg Med 2001; 19:337 395. Mabry B, Greller HA, Nelson LS. (2004). Patt ern of heroin overdose-induced pulmonary edema. 22(40:316 Mabry B, Greller HA, Nelson Ls. Pattern of Heroin Overdose-induced Pulmonary Edema. Am. J Emerg. Med. 2004; 22 (44):316. Matthias Wjst, Janine Altmller, Theresia Faus-Kessler, Christine Braig, Margret Bahnweg and Elisabeth Andr. (2006). Asthma families show transmission isequilibrium of gene variants in the vitamin D metabolism and signalling pathway Respiratory Research, 7:60 do i:10.11 McClure GM. (2000). Changes in suicid e in England and Wales 1960 1997. Br J Psychiatry; 176:64 67. 82
Monteis Catot, jaume, Evoluci on de las urgencias toxicologicas en un hospital general, M.Med., Universitat Autonoma de Ba rcelona (Spain), ProQuest, 1990, 214 pages; AAT C269144 Multnomah Heroin Overdose Deaths Multnomah County, Oregon, 1993-1999 MMWR, Morbidity and Mortality weekly report, CDC, July 21, 2000/vol.49/No 28) Myers Robert P, MD; Bing Li MA; A bdel Aziz M. Shaheen, MD, MPH. (2007). Emergency departement visits for acute acetaminophen overdose: a Canadianpopulation based epidemiologic study (1997-2002). Can J Emerg Med. 9(4): 267-74 National Institute on Drug Abuse (NIDA) InfoFacts. Drug addiction treatment methods.Available at: www.drugabuse.gov/infofax/treatmeth.html. Accessed March 19, 2008. National Institute on Drug Abuse (NIDA) InfoFacts. High school and youth trends. Available at: www.nida.nih.gov/InfoFax/HSYouthtrends.html. Accessed March 19,2008 National Institute on Drug Abuse (NIDA) InfoFacts. LSD. Available at: www.nida.nih.gov/Infofax/lsd.html. Accessed april 1, 2008. National Institute on Drug Abuse (NIDA) InfoFacts. Marijuana. Available a t: http:// National Institute on Drug Abuse (NIDA) InfoFacts. Marijuana. Available at: http://www.nida.nih.gov/Infofax/marijuana.html. Accessed March19, 2008. Nkhoma ET, Ed Hsu C, Hunt VI, Harris AM. Detecting spatiotemporal clusters of accidental poisoning mortality among Texas counties, U.S., 1980 2001. Int J Health geogr 2004; 3(1):25. 83
Nordentof Merete t, Leif breum, Lars K Munsk, Aksel G Nordestgaard, Alex Hunding, Peder A Laursen Bjoeldager.(1993) Hight mortality by natural and unnatural causes: a 10 years follow up st udy of patients admitted to a poisoning treatment center after suicide a ttempts. BMJ, 306:1637-41 Nordentoft, Merete: Prevention of suicide and attempted suicide in Denmark. (2007) Epidemiological studies of suicide and intervention studies in selected risk groups. danish medical bulletin. 54( 4):306-69 Nordentoft, Merete: Prevention of suicid e and attempted suicide in Denmark.(2007). Epidemiological studies of suicide and intervention studies in selected risk groups. Danish Medical Bulletin, 54(4) :306-69 Osborn HH, Tang M, Bradley K, Duncan BR, (1997) New-onset bronchospasm or recrudescence of asthma associated with cocaine abuse. Acad Energ Med. Jul; 4 (7): 689-92. Osborn HH, Tang M., Bradly K., Dunka n BR., New-onset bronchospasm or recrudescence of Asthma associated w ith cocaine abuse. Acad. Emerg Med. 1997 Jul; (7): 689-692 Osborn HH, Tang M., Bradly K., Dunka n BR., New-Onset bronchospasm or recrudescence of Asthma associated w ith cocaine abuse. Acad. Emerg Med. 1997 Jul; (7): 689-692 Osborn HH, Tang M., Bradly K., Dunka n BR., New-Onset bronchospasm or recrudescence of Asthma associated w ith cocaine abuse. Acad. Emerg Med. 1997; (7): 689-692 84
Price, D. J., Muro-Cacho, C. A., and Harbis on, R. D. (1999). The Role of Apoptosis in Phenobarbital Modulation of Cocaine -Induced Liver Damage. Research Communications in Alcohol and Substances of Abuse 20, 27 40 Psychiatr. Psychiatr Epidemiol 2001; 36(5):228 234. Quddusi Saista, Browne Patrick, Toivola Bert, (1998). Cushing syndrome due to superstitious glucocorticoid administration. Archive of Internal Medicine; 158(3):294-6. Robert S. McInnes, MPH & Dwight M. Bissell, MD MSPH. (1960). Accidental Poisoning and indication of hight accident frequency. Public Health reports 75(9):853-858. S. Hughes M.D, PMA Calverly MRCP: Hero in Inhalation and Asthma BMJ 1988; 297: 1611-1612. S. Hughes,& P M A Calverley. (1988). Hero in inhalation and asthma. BMJ 297: 15111512. Schauben JL and Rose Sr. Opiate Overdose In Aaron Ck and Linden CH, Eds. Intensive Care Medicine. 2nd ed. Boston: little, brown and Co., 1991:1232-1241. Seller E. A., MD., Ph.D., (1962) Unexpect ed hazards associated with drugs and chemicahls. The Canadian Medical asso ciation Journal 86(16), 721-724. Spillane JP, SchabenJL, Li S, Normann Sa, Wears R, Rose SR and augenstein WL. (1991). A retrospective Evaluation of Poisoning Exposure in Infants 0-12 Months of age. American Associatio n of Poisoning Control Centers annual Meeting, Atlanta, GA. 85
Spyros Papiris, Anastasia Katanidou, Katherin e Malagari & C. Rou ssos. (2002) Clinical review: severe asthma. Crit Care; 6(1): 30-44. Steensen P, Jorgensen HS & Juhl B. (1993). Heroin induced pulmonary edema. Ugeskr Laeger, 155(37): 2866-8 Steensen P, Jorgensen HS, juhl B.; Heroin -Induced Pulmonary edema Ugeskr laeger, 1993 sep 13; 165(37): 2866-8 Stein MD, O'Sullivan PS, Ellis P, et al. Utilization of medical services by drug abusers in detoxification. J Subst Abuse 1993; 5:187-93. Stein MD, O'Sullivan PS, Ellis P, et al. Utilization of medical services by drug abusers in detoxification. J. Subst Abuse 1993; 5:187-93. Stephens RS, Roffman RA, Curtin L. Compar ison of extended versus brief treatments for marijuana use. J Consult Clin Psychol 2000;68:898-908. Townsend E, Hawton K, Harriss L, et al. Substances used in delib erate self-poisoning 1985 1997: trends and associ ations with age, gender, repetition and suicide intent. Soc Psychiatr; Psychiatr Epidemiol 2001; 36(5):228 234. Thompson PJ, Dhillon P, Cole P. (1983). Addiction to aerosol treatment: the asthmatic alternative to glue sniffing. Br Med J (Clin Res Ed). 287(6404):1515-6. Townsend E, Hawton K, Harriss L, et al. Substances used in delib erate self-poisoning: trends and associations w ith age, gender, repetition and suicide intent, 1985 1997. Soc Psychiatry Psychiatr Epidemiol 2001; 36:228 234 86
Volkow, Nora D., M.D. Confronting the Rise in Abuse of Prescription Drugs National Institute on Drug Abuse (NIDA notes (2004/2005) 19(5): 2-3 Accessed March 19, 2008. http://www.nida.nih.gov/PDF/nidanotes/ NNvol19N5.pdf Volicer Beverly J., Quattrochi Nancy, Cande lieri Rose, Nicholosi Robert, Ladebauche Pauline. (2006). Depression and alcohol abuse in asthmatic college students. Nurse Practitioner; 31(2):49-54. Volkow ND. Confronting the rise in abuse of prescription drugs. NIDA notes 2005;19(5). Waldbott, GL, Asthma due to local anesthetics. JAMA;99, 1932; 1942 Weitzman JB; Kanarek NF; Smialek JE. ( 1998). Medical examiner asthma death autopsies: a distinct subg roup of asthma deaths with implications for public health preventive strategies. Arch Pathol Lab Med; 122(8):691-9. Withington, DE, Patrick, JA, Raynolds, F. (1993). Histamine release by morphine and dimorphine in man. Anesthesia 48:26-29 Xuei X, Flury-Wetherill L, Bierut L, Dick D, Nurnberger J Jr, Foroud T, Edenberg HJ. (2007). The opioid system in alcohol and drug dependence: family-based association study. Am J Med Genet B Neuropsychiatr Genet;144(7):877-84 Yurmez Yusuf, Polat Durukan, Yucel Yavuz, Ibra him Ikizceli, levent Avsarogullari, seda Ozkan, Okan Akudur, & Caglar Ozde mir. (2007). Acute organophosphate poisoning in university hospital emergency room patients, Internal Medicine. 46.6304: 965-969. 87
FINANCIAL DISCLOSURE The author declares that he has no conflict of interest with any third party organization or corporate involved on this project. 88
APPENDIX A: MODELS CALCULATION TABLES TABLE 6: Model Summary for Poisoning Final Outcome Model Sum of Squares df Mean Square F Sig. 6 Regression 19.545 6 3.257 14.796 .000f Residual 131.655 598 .220 Total 151.200 604 a. Predictors: (Constant), Cocaine b. Predictors: (Constant), Cocaine, INHALATION c. Predictors: (Constant), Co caine, INHALATION, P_POISON_HIS d. Predictors: (Constant), Cocaine, INHALATION, P_POISON_HIS, CVD2 e. Predictors: (Constant), Cocaine, IN HALATION, P_POISON_HIS, CVD2, ASTHMA2 f. Predictors: (Constant), Cocaine, INHALATION, P_POISON_HIS, CVD2, ASTHMA2, Gender g. Dependent Variable: OUTC_major 90
TABLE 6 (Continued): Model Summary for Poisoning Final Outcome Coefficients Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 6 (Constant) .108 .070 1.552 .121 Cocaine .200 .047 .186 4.283 .000 INHALATION .114 .046 .111 2.489 .013 P_POISON_HIS .090 .041 .089 2.215 .027 CVD2 .121 .052 .092 2.336 .020 ASTHMA2 .103 .042 .101 2.466 .014 Gender .094 .039 .094 2.401 .017 Model for Final Outcome of Poisoning = 0.200*Cocaine + 0.114 Inhalation + 0.121 CVD + 0.103 Asthma + 0.94 Gender + 0.09 Prior Po isoning History 91
TABLE 7 : Stepwise Regression Model Summ ary for Suicidal Predictability. Model Summary Mode l R R Square Adjusted R Square Std. Error of the Estimate Change Statistics R Square Change F Change df1 df2 Sig. F Change 1 .453a .206 .204 .304 .206 161.427 1 624 .000 2 .505b .255 .253 .295 .050 41.508 1 623 .000 3 .519c .269 .266 .292 .014 11.931 1 622 .001 4 .534d .285 .281 .289 .016 13.884 1 621 .000 5 .540e .291 .286 .288 .006 5.498 1 620 .019 6 .545f .297 .290 .287 .006 5.085 1 619 .024 Coefficients Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 6 (Constant) .394 .074 5.285 .000 DEPRESSION2 .330 .028 .398 11.617 .000 Age -.005 .001 -.195 -5.656 .000 Gender -.099 .024 -.144 -4.149 .000 PP.doc.his .072 .024 .104 2.962 .003 ASTHMA2 -.055 .024 -.079 -2.256 .024 unemployed -.082 .036 -.077 -2.255 .024 Predictors Model (Suicide) = 0.394 + 0.330 *Depression 0.005* Age -0.99* Gender + 0.072* Prior poisoning history 0.055 -0.082*unemployed. 92
TABLE 8: Heroin poisoning Predictors Model Summary Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate Change Statistics R Square Change F Change df1 df2 Sig. F Change 1 .351a .123 .122 .449 .123 87.441 1 624 .000 2 .423b .179 .176 .434 .056 42.557 1 623 .000 3 .461c .212 .209 .426 .033 26.366 1 622 .000 4 .500d .250 .245 .416 .037 30.942 1 621 .000 5 .523e .273 .267 .410 .023 19.940 1 620 .000 6 .531f .281 .274 .408 .008 7.152 1 619 .008 7 .537g .288 .280 .406 .007 5.968 1 618 .015 8 .542h .294 .285 .405 .005 4.800 1 617 .029 TABLE 8: (Continued) a. Predictors: (Constant), RECTAL_VAGINAL b. Predictors: (Constant), RECTAL_VAGINAL, ASTHMA2 c. Predictors: (Constant), RECTAL_VAGINAL, ASTHMA2, INHALATION d. Predictors: (Constant), RECTAL_VAGINAL, ASTHMA2, INHALATION, INJECTION e. Predictors: (Constant), RECTAL_VAGINAL, ASTHMA2, INHALATION, INJECTION, CVD2 f. Predictors: (Constant), RECTAL_VAGINAL, ASTHMA2, INHALATION, INJECTION, CVD2, PP.doc.his g. Predictors: (Constant), RECTAL_VAGINAL, ASTHMA 2, INHALATION, INJECTION, CVD2, PP.doc.his, h. Predictors: (Constant), RECTAL_VAGINAL, ASTH MA2, INHALATION, INJECTION, CVD2, PP.doc.his, Dependent Variable: Heroin 93
TABLE 8 (Continued): Heroin poisoning Predictors ANOVA and Model Coefficients ANOVAi Model Sum of Squares df Mean Square F Sig. 8 Regression 42.078 8 5.260 32.087 .000h Residual 101.141 617 .164 Total 143.220 625 Model Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 8 (Constant) .284 .079 3.576 .000 RECTAL_VAGIN AL .793 .061 .511 13.019 .000 ASTHMA2 .142 .035 .146 4.060 .000 INHALATION .273 .043 .278 6.414 .000 INJECTION .394 .080 .183 4.896 .000 CVD2 -.193 .043 -.153 -4.484 .000 PP.doc.his -.110 .035 -.113 -3.110 .002 ETOH2 -.077 .033 -.080 -2.306 .021 PSYCH2 -.111 .051 -.075 -2.191 .029 a. Dependent Variable: Heroin 94
Model (Heroin) = 0.284 + 0.793* Rectal/V aginal route + 0.142* Asthma + 0.273 Inhalation + 0.394*Injection 0.193* CVD 0.110* Prior documented Poisoning history 0.077* Alcohol 0.111 Psychosis TABLE 9: Asthma Model Predictors. A NOVA Model Sum of Squares df Mean Square F Sig. Regression 28.212 10 2.821 14.007 .000l Residual 124.674 619 .201 Total 152.886 629 a. Predictors: (Constant), INHALATION b. Predictors: (Constant), INHALATION, P_POISON_HIS c. Predictors: (Constant), INHALATION, P_POISON_HIS, Cocaine d. Predictors: (Constant), INHALATION, P_POISON_HIS, Cocaine, Gender e. Predictors: (Constant), INHALATION, P_POISON_HIS, Cocaine, Gender, RACE_AA f. Predictors: (Constant), INHALATION, P_POISON_HIS, Cocaine, Gender, RACE_AA, Hist.D.A g. Predictors: (Constant), INHALATION, P_POISON_HIS, Cocaine, Gender, RACE_AA, Hist.D.A, DERMAT h. Predictors: (Constant), INHALATION, P_POISON_HIS, Cocaine, Gender, RACE_AA, Hist.D.A, DERMAT, Age i. Predictors: (Constant), INHALATION, P_POISON_HIS, Cocaine, Gender, RACE_AA, Hist.D.A, DERMAT, Age, DIABETES j. Predictors: (Constant), INHALATION, P_POISON_HIS, Cocaine, Gender, RACE_AA, Hist.D.A, DERMAT, Age, DIABETES2, Heroin k. Predictors: (Constant), INHALATION, P_POISON_HIS, Gender, RACE_AA, Hist.D.A, DERMAT, Age, DIABETES2, Heroin l. Predictors: (Constant), INHALATION, P_POISON_HIS, Gender, RACE_AA, Hist.D.A, DERMAT, Age, DIABETES2, Heroin, RECTAL_VAGINAL m. Dependent Variable: ASTHMA2 95
TABLE 10 : Asthma Regressions Model Summary. Model Summary Mode l R R Square Adjusted R Square Std. Error of the Estimate 1 .242a .059 .057 .480 2 .304b .092 .090 .472 3 .336c .113 .109 .467 4 .365d .133 .128 .462 5 .390e .152 .145 .457 6 .401f .161 .153 .455 7 .411g .169 .160 .453 96
TABLE 10 (Continued): Asthma Regression Model Summary. a. Predictors: (Constant), INHALATION b. Predictors: (Constant), INHALATION, P_POISON_HIS c. Predictors: (Constant), INHALATION, P_POISON_HIS, Gender d. Predictors: (Constant), INHALATION, P_POISON_HIS, Gender, Heroin e. Predictors: (Constant), INHALATION, P_POISON_HIS, Gender, Heroin, CANCER2 f. Predictors: (Constant), INHALATION, P_POISON_HIS, Gender, Heroin, CANCER2, RACE_AA g. Predictors: (Constant), INHALATION, P_POISON_HIS, Gender, Heroin, CANCER2, RACE_AA, Age 97
98 TABLE 10 (Continued) Asthma Regression Models coeifficients. h. Dependent Variable: ASTHMA Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta (Constant) .419 .081 5.158 .000 INHALATIO N .221 .040 .217 5.579 .000 P_POISON_H IS .147 .039 .147 3.815 .000 Gender -.140 .037 -.141 -3.772 .000 Heroin .149 .039 .143 3.841 .000 CANCER2 .728 .187 .143 3.884 .000 RACE_AA .136 .045 .114 2.994 .003 Age -.004 .001 -.096 -2.481 .013 a. Dependent Variable: ASTHMA2
End Page ABOUT THE AUTHOR Dr. Khlifi A. Malek graduated from Monastir medical school in T unisia, he practiced medicine for four years and provided both cl inical and administrative role in a very prestigious organization in Tuni s capital. He then decided to move to the States, and soon after his arrival he enro lled into a Ph.D. program at the University of South Florida. He successfully passed all his me dical board exams in 2001 and got certified by ECFMG. He then accepted a job as a Pr ogram coordinator of a UN funded project in Saudi Arabia to assess the health damage s from the 1991 Gulf war as part of the oil for food program and he coauthored the United Nations Compensation CommissionÂ’s health claims report Â“F4Â” for compensation of Saudi Arabia on health. Dr. Khlifi was then hired as director of a legal and risk management department in a tertiary hospital in the Persian Gulf. He then returned from his overseas assignments to the states to pursue his career in toxicology.