Go to the Table of Contents

Skip to Main Content

OFFICE OF APPLIED STUDIES



Drug Abuse Warning Network, 2007:
National Estimates of Drug-Related Emergency Department Visits



U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES
Substance Abuse and Mental Health Services Administration
Office of Applied Studies



ACKNOWLEDGMENTS

This report was prepared by the Office of Applied Studies (OAS), Substance Abuse and Mental Health Services Administration (SAMHSA), and by RTI International (a trade name of Research Triangle Institute, Research Triangle Park, NC). Work by RTI was performed under Contract No. 280-03-2602.

PUBLIC DOMAIN NOTICE

All material appearing in this publication is in the public domain and may be reproduced or copied without permission from SAMHSA. Citation of the source is appreciated. However, this publication may not be reproduced or distributed for a fee without the specific, written authorization of the Office of Communications, SAMHSA, U.S. Department of Health and Human Services.

RECOMMENDED CITATION

Substance Abuse and Mental Health Services Administration, Office of Applied Studies. Drug Abuse Warning Network, 2007: National Estimates of Drug-Related Emergency Department Visits. Rockville, MD, 2010.

ELECTRONIC ACCESS AND COPIES OF PUBLICATION

This publication may be downloaded from http://DAWNinfo.samhsa.gov or from http://oas.samhsa.gov.

Or please call SAMHSA's Health Information Network at

1-877-SAMHSA-7 (1-877-726-4727)
(English and Español).

ORIGINATING OFFICE

Office of Applied Studies
Substance Abuse and Mental Health Services Administration
1 Choke Cherry Road, Rockville, MD 20857

May 2010



CONTENTS

Acknowledgments

Highlights
Drug misuse or abuse
Illicit drugs
Drugs and alcohol taken together
Alcohol use by patients aged 12 to 17 and 18 to 20
Nonmedical use of pharmaceuticals
Drug-related suicide attempts
Seeking detox services

Introduction
Major features of DAWN
What is a DAWN case?
What drugs are included in DAWN?
What is covered in this publication?
Hospital participation in 2007
Estimates in this publication
Margin of error for estimates
Comparisons across years
Rates of ED visits per 100,000 population
Limitations to data

Drug Misuse or Abuse
ED visits involving drug misuse or abuse, 2007
Trends in ED visits involving drug misuse and abuse, 2004–2007

Illicit Drugs
ED visits involving illicit drugs, 2007
Trends in ED visits involving illicit drugs, 2004–2007

Alcohol
ED visits involving drugs and alcohol taken together
ED visits involving alcohol use by patients aged 12 to 17 and 18 to 20
Trends in ED visits involving alcohol, 2004–2007

Nonmedical Use of Pharmaceuticals
ED visits involving nonmedical use of pharmaceuticals, 2007
Trends in ED visits involving nonmedical use of pharmaceuticals, 2004–2007

Drug-Related Suicide Attempts
ED visits involving drug-related suicide attempts, 2007
Trends in ED visits involving drug-related suicide attempts, 2004–2007

Seeking Detox Services
ED visits involving seeking detox services, 2007
Trends in ED visits involving seeking detox services, 2004–2007

List of Tables

Table 1. ED visits involving drug misuse and abuse, by drug combinations, 2007
Table 2. Trends in ED visits involving drug misuse and abuse, by drug combinations, 2004–2007
Table 3. ED visits involving illicit drugs, 2007
Table 4. Rates of ED visits per 100,000 population involving illicit drugs, 2007
Table 5. ED visits involving illicit drugs, by patient demographics, 2007
Table 6. Rates of ED visits per 100,000 population involving illicit drugs, by patient demographics, 2007
Table 7. ED visits and rates involving illicit drugs, by patient disposition, 2007
Table 8. Trends in ED visits involving illicit drugs, by selected drugs, 2004–2007
Table 9. ED visits involving alcohol, 2007
Table 10. ED visits involving drugs and alcohol taken together, by most frequent combinations, 2007
Table 11. ED visits involving drugs and alcohol taken together, by patient demographics, 2007
Table 12. ED visits involving drugs and alcohol taken together, by patient disposition, 2007
Table 13. ED visits involving alcohol, by presence of other drugs, by age groups 12 to 17 and 18 to 20, 2007
Table 14. ED visits involving alcohol, by presence of other drugs, 2004–2007
Table 15. ED visits involving alcohol, by presence of other drugs, by age groups 12 to 17 and 18 to 20, 2004–2007
Table 16. ED visits involving nonmedical use of pharmaceuticals, 2007
Table 17. ED visits and rates involving nonmedical use of pharmaceuticals, by patient demographics, 2007
Table 18. ED visits and rates involving nonmedical use of pharmaceuticals, by patient disposition, 2007
Table 19. Trends in ED visits involving nonmedical use of pharmaceuticals, by selected drugs, 2004–2007
Table 20. ED visits involving drug-related suicide attempts, by selected drugs, 2007
Table 21. ED visits involving drug-related suicide attempts, by patient demographics, 2007
Table 22. ED visits involving drug-related suicide attempts, by patient disposition, 2007
Table 23. Drug categories and drugs appearing more frequently in suicide attempts, 2005–2007
Table 24. Trends in ED visits for drug-related suicide attempts, by selected drugs, 2004–2007
Table 25. ED visits involving seeking detox services, by selected drugs, 2007
Table 26. ED visits involving seeking detox services, by patient demographics, 2007
Table 27. ED visits involving seeking detox services, by patient disposition, 2007
Table 28. Trends in ED visits involving seeking detox services, by selected drugs, 2004–2007

Table C1. Data collection year 2007
Table C2. Drug-related ED visits and drugs, by type of case, 2007
Table C3. DAWN analytic groups
Table C4. Oversample areas in DAWN sample design
Table C5. U.S. population by age and gender, 2007

Table D1. Drug-related ED visits, by detailed race/ethnicity, 2007

List of Figures

Figure 1. Rates of ED visits per 100,000 population involving illicit drugs, 2007
Figure 2. Rates of ED visits per 100,000 population involving illicit drugs, by selected drugs, age, and gender, 2007
Figure 3. Rates of ED visits per 100,000 population involving alcohol, by age and gender, 2007
Figure 4. Rates of ED visits per 100,000 population involving alcohol in combination and alcohol alone, by age groups 12 to 17 and 18 to 20, 2007
Figure 5. Rates of ED visits per 100,000 population involving nonmedical use of pharmaceuticals, by age and gender, 2007
Figure 6. Rates of ED visits per 100,000 population involving drug-related suicide attempts, by age and gender, 2007
Figure 7. Rates of ED visits per 100,000 population involving seeking detox services, by age and gender, 2007

Figure C1. DAWN ED case form
Figure C2. Type of case decision tree
Figure C3. Original DAWN sample design
Figure C4. DAWN design in practice

List of Appendices

Appendix A: Multum Lexicon End-User License Agreement
Appendix B: Glossary of Terms
Appendix C: DAWN Data Collection and Statistical Methods
Appendix D: Race and Ethnicity in DAWN


HIGHLIGHTS

This publication presents national estimates of drug-related visits to hospital emergency departments (EDs) for 2007, based on data from the Drug Abuse Warning Network (DAWN). Also presented are comparisons of 2007 estimates with those for 2004, 2005, and 2006. DAWN is a public health surveillance system that monitors drug-related ED visits for the Nation and for selected metropolitan areas. DAWN estimates pertain to the entire United States, including Alaska, Hawaii, and the District of Columbia. The Substance Abuse and Mental Health Services Administration (SAMHSA) is the agency responsible for DAWN. SAMHSA is required to collect data on drug-related ED visits under section 505 of the Public Health Service Act.

DAWN relies on a national sample of general, non-Federal hospitals operating 24-hour EDs. The sample is national in scope, with oversampling of hospitals in selected metropolitan areas. In each participating hospital, ED medical records are reviewed retrospectively to find the ED visits that involved recent drug use. All types of drugs—illegal drugs, prescription and over-the-counter pharmaceuticals (e.g., dietary supplements, cough medicine), and nonpharmaceutical inhalants—are included. Alcohol is considered a reportable drug when consumed by patients younger than 21. For patients aged 21 or older, though, alcohol is reported only when it is used in conjunction with other drugs.

Drug misuse or abuse

In 2007, hospitals in the United States delivered over 116 million ED visits, and DAWN estimates that about 1.9 million (1,883,272 [CI: 1,561,490 to 2,205,054])1 were associated with drug misuse or abuse.

Of the 1.9 million visits associated with drug misuse or abuse in 2007,

Although the overall number of ED visits attributable to drug misuse and abuse was stable from 2004 to 2007, ED visits involving nonmedical use of pharmaceuticals with no other drug involvement rose significantly (73%), as did the nonmedical use of pharmaceuticals with alcohol (36%).

Illicit drugs

For 2007, DAWN estimates that 974,272 (CI: 728,104 to 1,220,440) ED visits involved an illicit drug. That is, more than half (52%) of all the drug misuse/abuse ED visits during the year involved an illicit drug; multiple illicit drugs; or illicit drugs in combination with pharmaceuticals, alcohol, or both:

For each 100,000 persons in the U.S. population, over the course of a year, there are a little more than 180 ED visits (181.8 [CI: 125.7 to 237.9]) involving cocaine. This is followed by marijuana (101.3 ED visits per 100,000 persons), heroin (61.8), and stimulants (27.9). Lower incidence drugs have rates below 10 visits per 100,000 population. When the margin of error is taken into account, cocaine was involved significantly more often than any other illicit drug, and stimulants (amphetamines and methamphetamines) were involved less often than cocaine, marijuana, or heroin. Cocaine and heroin involvement were more common for patients aged 21 to 54 and less common for younger and older patients. Marijuana involvement was higher for patients aged 18 to 24, and stimulants involvement was higher for patients aged 18 to 44. Males were more likely than females to have a drug-related ED visit involving cocaine, marijuana, heroin, or stimulants. Just under half of the patients had some type of follow-up (i.e., referral to detox, admission to the hospital, or transfer to another facility).

Overall, the level of ED visits involving illicit drugs from 2004 to 2007 appeared stable for cocaine, marijuana, and heroin. The involvement of stimulants has declined consistently over that period, with about 75,000 fewer visits in 2007 than in 2004.

Drugs and alcohol taken together

The combination of drugs and alcohol is of particular concern because many illicit drugs and pharmaceuticals have additive or interactive effects with alcohol that can result in acute intoxication and impairment. In 2007, DAWN estimates 497,283 (CI: 406,698 to 587,868) ED visits, or 26 percent of all drug misuse or abuse visits, involved the use of alcohol in combination with other drugs. The drugs that alcohol was combined with most often included

The rate of ED visits for males (207.3 per 100,000 population) was higher than that for females (120.5 per 100,000 population). Rates by age group showed a general pattern of being lower for those younger than 18 or older than 54 and higher for those aged 18 to 54.

Almost half of the patients (49%) received some sort of follow-up treatment. Slightly more than a quarter (28%) of patients were admitted to the hospital, just under a fifth (14%) were transferred to another health care facility, and 8 percent were referred to a detox program.

From 2004 to 2007, no significant increases or decreases were found in the number of ED visits involving alcohol and other drugs. This finding was consistent for all age groups.

ED visits involving alcohol and no other drug are reportable to DAWN only if the patient is younger than 21. Consequently, these estimates do not represent the number of ED visits involving just alcohol among adults.

Alcohol use by patients aged 12 to 17 and 18 to 20

In 2007, almost 60,000 ED visits involved alcohol use with other drugs for patients aged 12 to 20, and more than 136,000 involved alcohol alone. More than two thirds (70%) of the total ED visits involving any type of alcohol use for this age group were associated with alcohol alone and no other drugs. Comparing patients aged 12 to 17 with those aged 18 to 20, the rates of ED visits per 100,000 population for use of alcohol alone were 223.2 and 619.9, respectively. The rates for the use of alcohol with other drugs were 105.3 and 249.5, respectively. Within both age groups, the rate of ED visits involving the use of alcohol alone is more than double that of alcohol plus drugs. Between age groups, the rate for patients aged 18 to 20 is more than double that of younger patients aged 12 to 17.

Looking across patients of all ages, no significant changes were found from 2004 to 2007 in the number of ED visits related to alcohol use alone in patients younger than 21. However, when the data are broken down by age groups, ED visits involving alcohol alone for patients aged 12 to 17 increased significantly (31%) from 2005 to 2007.

Nonmedical use of pharmaceuticals

For 2007, DAWN estimates that 855,838 (CI: 719,765 to 991,910) ED visits involved nonmedical use of prescription or over-the-counter pharmaceuticals or dietary supplements. Slightly more than half (52%) of these visits involved multiple drugs, and 19 percent involved alcohol. The rate of nonmedical use of pharmaceuticals did not differ between males and females. Most patients (53%) were treated and discharged home after their ED visit.

Central nervous system agents (present in 49% of nonmedical-use visits) and psychotherapeutic agents (41%) were the most frequent types of drugs reported in the nonmedical-use category of ED visits.

Among the central nervous system agents, the most frequent drugs were opiate/opioid pain medications (present in 34% of nonmedical-use visits). Methadone, oxycodone, and hydrocodone were the most frequently involved types of opioids.

Among the psychotherapeutic agents, the anxiolytics (anti-anxiety agents), sedatives, and hypnotics were the most frequent, occurring in almost a third (30%) of visits associated with nonmedical use of pharmaceuticals. ED visits involving benzodiazepines clearly outnumber those involving any of the other types of psychotherapeutic agents. DAWN estimates that 218,640 (CI: 179,649 to 257,632) ED visits, or 26 percent of nonmedical use of pharmaceuticals, involved benzodiazepines in 2007. Alprazolam was the most common type of benzodiazepine involved and was present in more than 80,000 visits.

ED visits involving nonmedical use of pharmaceuticals increased 60 percent in the period from 2004 to 2007. ED visits involving psychotherapeutic drugs, as an overarching drug category, increased 43 percent, a jump of more than 100,000 ED visits. Most types of psychotherapeutics—except antidepressants—saw significant increases in the period from 2004 to 2007. For example, benzodiazepines rose 52 percent, and central nervous system stimulants (a type of psychotherapeutic agent including amphetamine-dextroamphetamine and methylphenidate) rose 89 percent. ED visits involving central nervous system agents, as an overarching category, increased 47 percent, a jump of more than 130,000 ED visits. Most of those visits were related to a 66 percent increase in opiate/opioid pain medications.

Drug-related suicide attempts

DAWN estimates 197,053 (CI: 164,564 to 229,542) ED visits for drug-related suicide attempts in 2007.2 Females are more likely than males to be seen in the ED for a drug-related suicide attempt (78 visits per 100,000 population compared with 52). Rates are highest for those aged 18 to 20 (152 visits per 100,000 population). Almost all (94%) involved a pharmaceutical of some sort. More than half (57%) involved psychotherapeutic agents, nearly half (48%) involved central nervous system agents, just under a third (29%) involved alcohol, and about a fifth (19%) involved illicit drugs.3 Nearly two thirds (63%) of ED visits for drug-related suicide attempts involved multiple drugs.

While few patients were treated and discharged home (13%), most received some sort of follow-up care after the ED visits (e.g., transfer to another facility or admittance to the hospital intensive care unit [ICU/critical care], psychiatric unit, or other inpatient unit).

ED visits for drug-related suicide attempts increased 30 percent between 2005 and 2007. The rise in these visits reflects increases in visits related to psychotherapeutic agents (e.g., antidepressants) and central nervous system agents (e.g., painkillers). From 2005 to 2007, ED visits involving psychotherapeutic agents rose by about 30,000 visits, and visits involving central nervous system agents rose by about 28,000 visits. Specific drugs that stand out are hydrocodone (88% increase), benzodiazepines (50% increase), and acetaminophen (42% increase).

Seeking detox services

DAWN collects data on ED visits related to patients seeking detox. DAWN estimates 139,908 (CI: 110,901 to 168,915) drug-related ED visits in 2007 by patients seeking detoxification or substance abuse treatment services. Males are more likely than females to seek detox services through the ED (61 visits per 100,000 population compared with 32 visits).4

Two thirds (66%) of the seeking detox ED visits involved multiple drugs. Among illicit drugs, cocaine was observed in almost half of the visits (47%), heroin in about a third (30%), and marijuana in about a fifth (19%). Stimulants were relatively uncommon (5%). Among pharmaceuticals, benzodiazepines were observed in 14 percent of visits. Narcotic pain medications were observed in 27 percent of visits: oxycodone in 14 percent, hydrocodone in 8 percent, and methadone in 5 percent.

Among those seeking detox, nearly 7 out of 10 (69%) received some type of follow-up care. Nearly a third (30%) of patients were released with a referral to a detox or treatment program, about a quarter (22%) were admitted to the detox unit in the hospital, 8 percent were admitted to another unit in the hospital, and 9 percent were transferred to other facilities. A quarter of patients (25%) were treated and discharged home.

The number of patients seeking detox services through the ED was relatively stable from 2004 through 2007. Some significant changes were observed at the drug level, though. A significant increase (41%) was observed in seeking detox visits involving opiate/opioid pain medications from 2005 to 2007. Narcotic painkillers in general, and hydrocodone and oxycodone in particular, were a large part of the increased visits observed in 2007. There has been no significant increase overall in the number of seeking detox visits that involve psychotherapeutic agents (e.g., antidepressants, sedatives).


INTRODUCTION

This publication presents estimates of drug-related emergency department (ED) visits from the Drug Abuse Warning Network (DAWN) for 2007, with comparison of estimates for 2004, 2005, and 2006. DAWN is a public health surveillance system that monitors drug-related ED visits for the Nation and for selected metropolitan areas. The Office of Applied Studies (OAS) of the Substance Abuse and Mental Health Services Administration (SAMHSA), U.S. Department of Health and Human Services, has been responsible for DAWN operations since 1992.

This introduction provides a brief description of the major features of DAWN and the statistics presented in this report. Findings are organized in six sections following this Introduction. Each section focuses on a specific type of ED visit. Appendix B: Glossary of Terms and Appendix C: DAWN Data Collection and Statistical Methods provide additional detail on all aspects of the DAWN sample, data collected and data collection methodology, response rates, survey weights and adjustments, statistical processing, and terms used in this report.

Major features of DAWN

What is a DAWN case?

A DAWN case is any ED visit involving recent drug use that is implicated in the ED visit. The relationship between the ED visit and the drug use need not be causal. That is, implicated drugs may or may not have directly caused the condition generating the ED visit. The reason a patient used a drug is not part of the criteria for considering a visit to be drug related. These criteria broadly encompass all types of drug-related events, including accidental ingestion and adverse reaction, as well as explicit drug abuse. DAWN does not report current medications (i.e., medications and pharmaceuticals taken regularly by the patient as prescribed or indicated) that are unrelated to the ED visit.

What drugs are included in DAWN?

DAWN collects data on all types of drugs, including

What is covered in this publication?

This publication focuses primarily on ED visits involving drug misuse and abuse. Seven types of ED visits associated with drug misuse and abuse are highlighted in this publication:

Drug misuse and abuse is an overarching category that includes all drug-related ED visits involving drug misuse or abuse. Use of illicit drugs is singled out for analysis as it involves the use of substances that are generally illegal. By definition, this is substance abuse. Visits involving alcohol used in combination with other drugs are analyzed as a group to better understand the interactive effects of alcohol and drugs on morbidity. ED visits involving underage drinking (without use of other drugs) are an important barometer of dangerous drinking patterns in youths. Nonmedical use of pharmaceuticals concerns ED visits related to the misuse or abuse of prescription or over-the-counter medications or dietary supplements. This might result from taking a higher-than-prescribed or -recommended dose of a pharmaceutical (i.e., contrary to directions or labeling), taking a pharmaceutical prescribed for another individual, malicious poisoning of the patient by another individual, and substance abuse involving pharmaceuticals. Drug-related suicide attempts involve drug overdoses as well as suicide attempts by other means (e.g., by gun) if drugs were involved. "Seeking detox" includes various situations such as nonemergency requests for admission for detox, visits to obtain medical clearance before entry to a detox program, and acute emergencies in which an individual is in distress (i.e., displaying active withdrawal symptoms) and seeking detox.

Hospital participation in 2007

For 2007, 207 hospitals submitted data that were used for estimation. The overall weighted response rate was 29.6 percent. For the 12 oversampled metropolitan areas and divisions, the individual response rates ranged from 30.7 percent in the Houston metropolitan area to 76.3 percent in the Detroit metropolitan area.5

DAWN cases are found through a retrospective review of medical records in participating hospitals. Across all participating hospitals in 2007, 10.4 million charts were reviewed to find the drug-related ED visits that met the DAWN case criteria. On the basis of the review of charts, 375,030 drug-related visits were found and submitted. On average, a DAWN member hospital submitted 1,183 DAWN cases. However, the number of submitted cases varied widely across hospitals, from 3 cases to 6,532 cases (median 953) in a single hospital during 2007.

Estimates in this publication

The estimates provided in this publication represent drug-related ED visits for the United States. The universe of hospitals eligible for inclusion in DAWN includes non-Federal, short-stay, general medical and surgical hospitals in the United States that operate 24-hour EDs. The American Hospital Association's (AHA) 2001 Annual Survey was used to identify the original frame members. Subsequent AHA surveys are used annually to identify "births" of new hospitals that open and the "deaths" of hospitals that close or merge with other hospitals.

The DAWN sample of hospitals includes an oversampling of hospitals in selected metropolitan areas, supplemented with a sample of hospitals from the remainder of the United States, which includes other metropolitan areas as well as nonmetropolitan and rural areas. The metropolitan area boundaries correspond to the definitions issued by the Office of Management and Budget (OMB) in June 2003.

Estimates of drug-related ED visits are calculated by applying weights and adjustments to the data provided by the sampled hospitals participating in DAWN. The primary sampling weights reflect the probability of selection, and separate adjustment factors are included to account for nonresponse, data quality, and the known total of ED visits delivered by the universe of eligible hospitals as reported by the most current AHA survey.

DAWN currently collects drug information using more than 17,000 individual codes.6 These highly detailed codes are grouped up (mapped) to 3,200 drug names. Drug names are then grouped into 500 broader drug categories. About 100 of the more common drugs and drug categories were selected for inclusion in the drug detail tables in this report. Because a single ED visit may involve multiple drugs and the same drug may be reported both under its specific drug name and under its drug category, the sum of ED visits from different rows in the drug detail tables will be greater than the total number of visits. For the same reason, percentages will add to more than 100.

Margin of error for estimates

Because DAWN relies on a sample of hospitals, each estimate produced from the DAWN ED data is subject to sampling variability, referred to as the "margin of error." Margin of error is the variation in the estimate that would be observed naturally if different samples were drawn from the same population using the same procedures. The sampling variability of an estimate in this publication is measured by its relative standard error (RSE). The precision of an estimate is inversely related to its sampling variability, as measured by the RSE. That is, the greater the RSE, the lower the precision.

DAWN estimates with RSE values greater than 50 percent or fewer than 30 ED visits, or both, are regarded as too imprecise for publication and are not shown. An asterisk (*) is displayed in the place of suppressed estimates. Ratios (percentages or rates per 100,000 population) based on suppressed estimates are likewise suppressed.

In this publication, 95 percent confidence intervals (CIs) are included in many of the tables and are cited in the text along with the estimates. A 95 percent CI means that if repeated samples were drawn from the same population of hospitals using the same sampling and data collection procedures, the true population value would fall within the confidence interval 95 percent of the time. A CI, which is expressed as a range of values, is useful because the interval reflects both the estimate and its particular margin of error.

Comparisons across years

In this publication, between-year changes are assessed by comparing estimates for 2007 with those for 2004, 2005, and 2006. This publication reports only those between-year changes that are statistically significant at the p < 0.05 level.

Major changes to DAWN were instituted during 2003 as the result of a redesign that altered most of DAWN's core features. Changes included the design of the hospital sample, the drug-related cases eligible for DAWN, the data items submitted on these cases, and the protocol for case finding and quality assurance. These improvements created a permanent disruption in trends. As a result, comparisons cannot be made between old DAWN (2003 and prior years) and the redesigned DAWN (2004 and forward).

Rates of ED visits per 100,000 population

Standardized measures are helpful when comparing levels of drug-related ED visits for different age groups and genders. This publication reports rates of ED visits per 100,000 persons by age and gender. Rates are based on population data from the U.S. Census Bureau. If an estimate is suppressed, the rate will also be suppressed. Unfortunately, DAWN is unable to produce population-based rates for race/ethnicity categories. Race/ethnicity information in ED records is often missing or is very limited. Furthermore, DAWN uses a simplified set of race/ethnicity categories that is incompatible with the categories used by the U.S. Census Bureau to report population by race/ethnicity. Therefore, the population denominators that would enable DAWN to produce rates are not available.

Limitations to data

Information on drug-related visits is based on a sample and is, therefore, subject to sampling variability. Readers are advised to consider the standard error measurements provided in many tables to reflect the sampling variability that occurs by chance because only a sample rather than the entire universe is surveyed. Hospital participation rates in oversampled metropolitan areas typically have been 50 percent or higher. However, the participation rate in the remainder of the United States has been lower, in the range of 20 to 30 percent, since the DAWN redesign in 2003. In any sample survey, a low response rate is of concern because it creates the opportunity for bias. That is, nonparticipating hospitals may have different characteristics than participating hospitals, possibly including differences in the drugs reported, patient disposition, or population demographics. DAWN is addressing these issues by developing statistical and data collection methods that help to avoid or minimize bias and improve response rates within available resources.

Although every effort is made during the data collection phase to collect data accurately and precisely, extant medical records vary in specificity and detail. Therefore, factors that may affect the reliability and accuracy of the findings include the following:


DRUG MISUSE OR ABUSE

ED visits involving drug misuse or abuse, 2007

For 2007, DAWN estimates that almost 1.9 million ED visits were associated with drug misuse or abuse (Table 1). This estimate includes

Table 1
ED visits involving drug misuse and abuse, by drug combinations, 2007
Drug combinations (1) ED visits (2) Percent of ED visits RSE (%) 95% CI: Lower bound 95% CI: Upper bound
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(3) DAWN excludes ED visits involving alcohol-only for patients aged 21 years or older. Alcohol, when present with other drugs, is reportable for patients of all ages.
NOTE: CI = confidence interval. RSE = relative standard error. A dash (—) indicates a blank cell.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, drug misuse/abuse 1,883,272 100.0   8.7 1,561,490 2,205,054
Illicit drugs only    522,650   27.8 15.9    360,142    685,158
Alcohol only (age < 21)    137,369     7.3   8.4    114,635    160,103
Nonmedical use of pharmaceuticals only    582,187   30.9   9.0    479,464    684,909
Combinations           —     —    —           —           —
Illicit drugs with alcohol (3)    237,936   12.6 12.6    179,073    296,799
Illicit drugs with any pharmaceuticals    143,783     7.6 10.8    113,436    174,129
Alcohol with nonmedical use of pharmaceuticals    189,444   10.1   7.4    162,068    216,821
Illicit drugs with alcohol and any pharmaceuticals      69,903     3.7 13.2      51,877      87,929

Of the almost 1.9 million drug misuse/abuse visits, about two thirds (66%) were associated with a single drug type (illicit drugs, alcohol, or nonmedical use of pharmaceuticals). Illicit drugs alone were involved in 28 percent of drug misuse/abuse visits in 2007. Nonmedical use of pharmaceuticals alone was involved in 31 percent. About 7 percent of drug misuse/abuse visits involved consumption of alcohol (and no other drug) by a minor.8 The remaining visits (34%) involved some combination of illicit drugs, alcohol, and nonmedical use of pharmaceuticals.

These figures do not suggest that the majority of ED drug misuse/abuse visits involved a single drug. In fact, the typical drug-related ED visit involves multiple drugs, but they may be of a common type. For example, an ED visit involving illicit drugs alone often involves more than one illicit drug (e.g., cocaine and heroin).

Trends in ED visits involving drug misuse and abuse, 2004–2007

This section presents the trends in the estimates of ED visits involving drug misuse and abuse for the period 2004 through 2007 (Table 2). Differences between years are presented in terms of the percentage increase or decrease in visits in 2007 compared with the estimates for the previous 3 years. Only statistically significant changes are discussed and displayed in the table.

Table 2
Trends in ED visits involving drug misuse and abuse, by drug combinations, 2004–2007
Drug combinations (1) ED visits, 2004 (2) ED visits, 2005 (2) ED visits, 2006 (2) ED visits, 2007 (2) Percent change 2004, 2007 (3) Percent change 2005, 2007 (3) Percent change 2006, 2007 (3)
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(3) This column denotes statistically significant (p < 0.05) increases or decreases between estimates for the periods shown.
NOTE: A dash (—) indicates a blank cell.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
All types of drug misuse/abuse 1,619,054 1,616,311 1,742,887 1,883,272
Illicit drugs only    502,136    517,558    536,554    522,650
Alcohol only (age < 21)    150,988    110,599    126,704    137,369
Nonmedical use of pharmaceuticals only    336,987    444,309    486,276    582,187 73 20
Combinations            —            —            —            —
Illicit drugs with alcohol    338,638    221,823    219,521    237,936
Illicit drugs with pharmaceuticals    105,017    127,245    142,535    143,783
Alcohol with nonmedical use of pharmaceuticals    139,716    140,275    171,743    189,444 36 35
Illicit drugs with alcohol and pharmaceuticals      45,571      54,500      59,553      69,903

The number of ED visits attributable to drug misuse and abuse was stable from 2004 to 2007. The small changes seen in the estimates each year are within the boundaries of expected sample variation. From 2004 to 2007, however, ED visits related to the use of pharmaceuticals with no other drug involvement rose significantly (73%), as did the use of pharmaceuticals with alcohol (36%). These increases reflect nearly a quarter-million more ED visits related to pharmaceuticals alone and about 50,000 more ED visits related to pharmaceuticals and alcohol in 2007 than in 2004.

By way of comparison, hospitals in the United States delivered a total of more than 116 million ED visits in 2007, an increase of 7.0 percent over 2004. The population of the United States increased 3.9 percent, from 294 million to 304 million, over the same period.


ILLICIT DRUGS

ED visits involving illicit drugs, 2007

For 2007, DAWN estimates that 974,272 (CI: 728,104 to 1,220,440) ED visits involved one or more illicit drugs (Table 3). This represents about half (52%) of the approximately 1.9 million drug misuse and abuse ED visits that occurred during 2007. Cocaine was the most commonly involved drug, with 553,530 (CI: 382,646 to 724,414) ED visits. Nearly one in three drug misuse/abuse ED visits (29%) involved cocaine. Marijuana followed cocaine, with 308,547 (CI: 250,529 to 366,564) ED visits, or 16 percent. Heroin was involved in 188,162 (CI: 130,391 to 245,932) ED visits, or approximately 10 percent of drug misuse/abuse ED visits overall.9 Stimulants, including amphetamines and methamphetamine, were involved in 85,043 (CI: 48,398 to 121,689) ED visits, about 5 percent of drug misuse/abuse ED visits.

Table 3
ED visits involving illicit drugs, 2007
Drugs (1) ED visits (2,3) Percent of ED visits (3) RSE (%) 95% CI:
Lower bound
95% CI:
Upper bound
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(3) ED visits often involve multiple drugs. Such visits will appear multiple times in this table (e.g., a visit involving both cocaine and marijuana will appear twice in this table). The sum of visits or rates by drug will be greater than the total, and the sum of percentages by drug will be greater than 100.
NOTE: CI = confidence interval. RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, drug misuse/abuse 1,883,272 100.0   8.7 1,561,490 2,205,054
Total ED visits, illicit drugs    974,272   51.7 12.9    728,104 1,220,440
Visits involving a single illicit drug    425,833   22.6 16.6    287,025    564,641
Visits involving multiple drugs    548,439   29.1 11.3    427,160    669,718
Cocaine    553,530   29.4 15.8    382,646    724,414
Heroin    188,162   10.0 15.7    130,391    245,932
Marijuana    308,547   16.4   9.6    250,529    366,564
Stimulants      85,043     4.5 22.0      48,398    121,689
Amphetamines      21,545     1.1 17.4      14,194      28,896
Methamphetamine      67,954     3.6 25.3      34,266    101,641
MDMA (Ecstasy)      12,748     0.7 18.5        8,120      17,376
GHB        2,207     0.1 35.7           663        3,752
Flunitrazepam (Rohypnol)              *       *     *              *              *
Ketamine          291  <0.1 30.1           120           463
LSD        3,561     0.2 22.2        2,009        5,112
PCP      28,035     1.5 39.0        6,585      49,484
Miscellaneous hallucinogens        4,839     0.3 18.4        3,097        6,582
Inhalants        7,920     0.4 17.8        5,159      10,681
Combinations not tabulated above        3,989     0.2 28.4        1,766        6,213

Other illicit drugs involved in ED visits at lower levels include the following:

The rates of ED visits involving illicit drugs are reported in Table 4. For each 100,000 persons in the U.S. population, over the course of a year, more than 600 ED visits relate to drug misuse or abuse (618.5 [CI: 512.8 to 724.2]). About half of those visits involve illicit drugs. Cocaine is involved at a rate of 181.8 (CI: 125.7 to 237.9) ED visits per 100,000 persons in the United States, followed by marijuana (101.3 [CI: 82.3 to 120.4] ED visits per 100,000 persons), heroin (61.8 [CI: 42.8 to 80.8]), and stimulants (27.9 [CI: 15.9 to 40.0]). Lower-incidence drugs have rates below 10 persons per 100,000 population.

Table 4
Rates of ED visits per 100,000 population involving illicit drugs, 2007
Drugs (1) Rate of ED visits per 100,000 population (2,3) RSE (%) 95% CI:
Lower bound
95% CI:
Upper bound
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) All rates are ED visits per 100,000 population. Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States. Population estimates are drawn from the 2007 U.S. Census Bureau Postcensal Resident Population National Population Dataset as of July 1, 2007.
(3) ED visits often involve multiple drugs. Such visits will appear multiple times in this table (e.g., a visit involving both cocaine and marijuana will appear twice in this table). The sum of visits or rates by drug will be greater than the total, and the sum of percentages by drug will be greater than 100.
NOTE: CI = confidence interval. RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, drug misuse/abuse 618.5   8.7 512.8 724.2
Total ED visits, illicit drugs 320.0 12.9 239.1 400.8
Cocaine 181.8 15.8 125.7 237.9
Heroin   61.8 15.7   42.8   80.8
Marijuana 101.3   9.6   82.3 120.4
Stimulants   27.9 22.0   15.9   40.0
Amphetamines     7.1 17.4     4.7     9.5
Methamphetamine   22.3 25.3   11.3   33.4
MDMA (Ecstasy)     4.2 18.5     2.7     5.7
GHB     0.7 35.7     0.2     1.2
Flunitrazepam (Rohypnol)       *     *       *       *
Ketamine     0.1 30.1     0.0     0.2
LSD     1.2 22.2     0.7     1.7
PCP     9.2 39.0     2.2   16.3
Miscellaneous hallucinogens     1.6 18.4     1.0     2.2
Inhalants     2.6 17.8     1.7     3.5
Combinations not tabulated above     1.3 28.4     0.6     2.0

Figure 1 displays the rates of ED visits per 100,000 population for the four major types of illicit drugs: cocaine, marijuana, heroin, and stimulants.

Figure 1
Rates of ED visits per 100,000 population involving illicit drugs, 2007

Figure 1   D

SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).

Table 5 presents estimates of the number of ED visits in 2007 involving illicit drugs for males and females, different age groups, and race/ethnicity categories. To facilitate comparisons, Table 6 and Figure 2 present the rates of ED visits per 100,000 persons for these same groups. The rates for visits involving cocaine, marijuana, heroin, and stimulants were consistently higher for males than for females. As to the age of the patient, rates of ED visits involving illicit drugs are generally higher for those aged 18 to 54. Notable exceptions are the higher rates found for 12- to 17-year-olds for marijuana, MDMA (Ecstasy), and PCP. Within the 18-to-54 age group, rates of ED visits involving cocaine peak at about 434 visits per 100,000 persons for patients aged 35 to 44, with heroin and stimulants found at lower levels and more evenly dispersed across the 18-to-54 age range.

Table 5
ED visits involving illicit drugs, by patient demographics, 2007
Patient demographics All illicits (1) Cocaine Heroin Marijuana Stimulants MDMA (Ecstasy) GHB LSD PCP
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(3) ED visits often involve multiple drugs. Such visits will appear multiple times in this table (e.g., a visit involving both cocaine and marijuana will appear twice in this table). The sum of visits or rates by drug will be greater than the total, and the sum of percentages by drug will be greater than 100.
NOTE: RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed. A dash (—) indicates a blank cell.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, illicit drugs (2,3) 974,272 553,530 188,162 308,547 85,043 12,748 2,207 3,561 28,035
Gender         —         —         —         —       —       —     —     —       —
Male 639,799 352,320 132,662 205,308 54,025   7,607 1,994 2,742 20,528
Female 334,010 201,008   55,398 103,111 30,953   5,141    213    819   7,507
Unknown        463           *           *           *         *         *       *       *         *
Age         —         —         —         —       —       —     —     —       —
0–5 years     1,059        172           *           *         *         *       *       *         *
6–11 years        538           *           *           *         *         *       *       *         *
12–17 years   61,074     9,314     2,322   50,126   5,125   2,226       *    263   1,818
18–20 years   73,223   22,892   10,765   43,280   7,303   2,631       * 1,117         *
21–24 years 101,436   40,307   20,307   45,823 14,321   3,484       * 1,247   3,023
25–29 years 127,382   64,294   28,466   45,485 13,382   2,368       *    504   6,469
30–34 years 111,003   65,635   22,448   33,241 13,006   1,218       *    172   5,477
35–44 years 259,469 184,899   49,079   52,707 19,253      705       *    126   5,871
45–54 years 188,596 135,231   40,882   28,159 10,528        86       *       *   1,883
55–64 years   42,714   26,310   11,991     7,365   1,282         *       *       *         *
65 years and older     7,442     4,262     1,755     1,867         *         *       *       *         *
Unknown        337        157           *           *         *         *       *       *         *
Race/ethnicity         —         —         —         —       —       —     —     —       —
White 445,031 214,380   91,300 165,923 50,182   5,977 1,427 2,551   8,521
Black 307,860 225,923   40,751   80,774   6,033   3,192       *      76 14,761
Hispanic 122,055   61,108   32,764   35,112         *   1,453       *    264    2,792
Other or two or more race/ethnicities     9,636     3,992     1,595     2,613   1,956      514       *       *         *
Unknown   89,689   48,126   21,752   24,125 12,111   1,611       *    374   1,687
Table 6
Rates of ED visits per 100,000 population involving illicit drugs, by patient demographics, 2007
Patient demographics All illicits (1) Cocaine Heroin Marijuana Stimulants MDMA (Ecstasy) GHB LSD PCP
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) All rates are ED visits per 100,000 population. Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States. Population estimates are drawn from the 2007 U.S. Census Bureau Postcensal Resident Population National Population Dataset as of July 1, 2007.
(3) ED visits often involve multiple drugs. Such visits will appear multiple times in this table (e.g., a visit involving both cocaine and marijuana will appear twice in this table). The sum of visits or rates by drug will be greater than the total, and the sum of percentages by drug will be greater than 100.
NOTE: RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed. A dash (—) indicates a blank cell. Rates are not provided for race and ethnicity subgroups because of data limitations.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Rates of ED visits, illicit drugs (2,3) 320.0 181.8   61.8 101.3 27.9   4.2 0.7 1.2   9.2
Gender      —      —      —      —    —    —    —
Male 426.1 234.6   88.3 136.7 36.0   5.1 1.3 1.8 13.7
Female 216.4 130.3   35.9   66.8 20.1   3.3 0.1 0.5   4.9
Age      —      —      —      —    —    —    —
0–5 years     4.2     0.7        *        *      *      *    *    *      *
6–11 years     2.3        *        *        *      *      *    *    *      *
12–17 years 243.6   37.2     9.3 199.9 20.4   8.9    * 1.0   7.3
18–20 years 565.6 176.8   83.1 334.3 56.4 20.3    * 8.6      *
21–24 years 599.7 238.3 120.1 270.9 84.7 20.6    * 7.4 17.9
25–29 years 593.5 299.6 132.6 211.9 62.3 11.0    * 2.3 30.1
30–34 years 563.1 333.0 113.9 168.6 66.0   6.2    * 0.9 27.8
35–44 years 608.9 433.9 115.2 123.7 45.2   1.7    * 0.3 13.8
45–54 years 424.9 304.7   92.1   63.4 23.7   0.2    *    *   4.2
55–64 years 126.9   78.1   35.6   21.9   3.8      *    *    *      *
65 years and older   19.2   11.0     4.5     4.8      *      *    *    *      *

Figure 2
Rates of ED visits per 100,000 population involving illicit drugs, by selected drugs, age, and gender, 2007

Figure 2   D

SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).

Estimates of ED visits related to illicit drugs reveal that 46 percent of patients are white, 32 percent are black, 13 percent are Hispanic, 1 percent are other or multiple races/ethnicities, and 9 percent are of unknown race/ethnicity. Unfortunately, DAWN is unable to produce population-based rates for race/ethnicity categories. Race/ethnicity information in ED records is often missing or is very limited. By necessity, DAWN uses a simplified set of race/ethnicity categories that is incompatible with the categories used by the U.S. Census Bureau to report population by race/ethnicity. Therefore, the population denominators that would enable DAWN to produce rates are not available.

EDs are a potential site to intercept patients for follow-up treatment for drug use problems. Table 7 displays patient disposition after ED visits involving illicit drugs. A majority of patients (57%) are treated and released, about a quarter (26%) are admitted to the hospital, and the balance (17%) have other outcomes.

Table 7
ED visits and rates involving illicit drugs, by patient disposition, 2007
Patient disposition ED visits (1) Percent of ED visits Rate of ED visits per 100,000 population (2)
(1) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(2) All rates are ED visits per 100,000 population. Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States. Population estimates are drawn from the 2007 U.S. Census Bureau Postcensal Resident Population National Population Dataset as of July 1, 2007.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, illicit drugs 974,272 100.0 320.0
Treated and released 560,065   57.5 183.9
Discharged home 426,581   43.8 140.1
Released to police/jail   55,670     5.7   18.3
Referred to detox/treatment   77,814     8.0   25.6
Admitted to this hospital 253,388   26.0   83.2
ICU/critical care   35,165     3.6   11.5
Surgery     2,534     0.3     0.8
Chemical dependency/detox   32,720     3.4   10.7
Psychiatric unit   65,534     6.7   21.5
Other inpatient unit 117,435   12.1   38.6
Other follow-up 160,818   16.5   52.8
Transferred 105,428   10.8   34.6
Left against medical advice   22,518     2.3     7.4
Died     1,658     0.2     0.5
Other   22,710     2.3     7.5
Not documented     8,505     0.9     2.8

Trends in ED visits involving illicit drugs, 2004–2007

This section presents the trends in the estimates of ED visits involving illicit drugs for the period 2004 through 2007 (Table 8). Differences between years are presented in terms of the percentage increase or decrease in visits in 2007 compared with the estimates for the previous 3 years. Only statistically significant changes are discussed and displayed in the table.

Table 8
Trends in ED visits involving illicit drugs, by selected drugs, 2004–2007
Drugs (1) ED visits, 2004 (2,3) ED visits, 2005 (2,3) ED visits, 2006 (2,3) ED visits, 2007 (2,3) Percent change 2004, 2007 (4) Percent change 2005, 2007 (4) Percent change 2006, 2007 (4)
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(3) ED visits often involve multiple drugs. Such visits will appear multiple times in this table (e.g., a visit involving both cocaine and marijuana will appear twice in this table). Thus, the sum of visits or rates by drug will be greater than the total, and the sum of percentages by drug will be greater than 100.
(4) This column denotes statistically significant (p < 0.05) increases or decreases between estimates for the periods shown.
NOTE: RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed. A dash (—) indicates a blank cell.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, drug misuse/abuse 1,619,054 1,616,311 1,742,887 1,883,272
ED visits, illicit drugs    991,363    921,127    958,164    974,272
Cocaine    475,425    483,865    548,608    553,530
Heroin    214,432    187,493    189,780    188,162
Marijuana    281,619    279,664    290,563    308,547
Stimulants    162,435    137,650    107,575      85,043 -38 -21
Amphetamines      34,085      34,928      32,240      21,545 -38 -33
Methamphetamine    132,576    109,655      79,924      67,954 -38 -15
MDMA (Ecstasy)      10,220      11,287      16,749      12,748 -24
GHB        1,789        1,036        1,084        2,207
Flunitrazepam (Rohypnol)              *              *              *              *
Ketamine              *           303           270           291
LSD        2,146        2,001        4,002        3,561
PCP      31,342      14,825      21,960      28,035  28
Miscellaneous hallucinogens        3,150        3,194        3,898        4,839
Inhalants        9,523        5,156        5,643        7,920
Combinations not tabulated above              *        3,201        2,055        3,989


Overall, the level of ED visits involving illicit drugs from 2004 to 2007 appeared stable, with no significant changes from 2004 to 2007 for three of the four major illicit drugs: cocaine, marijuana, and heroin. The fourth major illicit drug category, stimulants, has declined consistently since 2004, with about 75,000 fewer visits in 2007 than in 2004. Among the illicit drugs found at lower levels, PCP increased and MDMA (Ecstasy) declined from 2006 to 2007, resulting in about 6,000 more ED visits related to PCP and 4,000 fewer visits for MDMA. More years' data are needed to know if these changes signify trends.

ALCOHOL

In 2007, almost a half-million ED visits involved drugs combined with alcohol (Table 9). This represented more than a quarter (26%) of all the ED visits involving drug misuse/abuse that year. The combination of drugs and alcohol is of particular concern because many illicit drugs and pharmaceuticals have additive or interactive effects with alcohol that can result in acute intoxication and impairment. According to the National Institute on Alcohol Abuse and Alcoholism (NIAAA), more than 150 medications interact harmfully with alcohol. These interactions may result in increased risk of illness, injury, and even death. Alcohol's effects are heightened by drugs that depress the central nervous system, such as heroin, opiate pain medications, benzodiazepines, antihistamines, and antidepressants. Medications for certain disorders, including diabetes, high blood pressure, and heart disease, also can have harmful interactions with alcohol.10

Table 9
ED visits involving alcohol, 2007
Alcohol use category (1) ED visits (2) Percent of ED visits RSE (%) 95% CI:
Lower bound
95% CI:
Upper bound
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
NOTE: CI = confidence interval. RSE = relative standard error.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Alcohol with drugs (all ages) 497,283 26.4 9.3 406,698 587,868
Alcohol alone (patients < 21) 137,369   7.3 8.4 114,635 160,103

The use of alcohol alone by those under age 21 is also of substantial concern. Underage drinking has many immediate adverse consequences, and it also can lead to higher levels and dangerous patterns of drinking in later years. As an indicator of the prevalence and severity of underage drinking, its consequences, and its trends through the teen years, DAWN reports on ED visits for underage drinking separately for adolescents aged 12 to 17 and 18 to 20.

ED visits involving drugs and alcohol taken together

The types of drugs that accompany alcohol use are displayed in Table 10. Illicit drugs were involved in well over half (62%) of ED visits involving alcohol and other drugs. One or more pharmaceuticals were involved in 52 percent of the visits. Psychotherapeutic agents such as antidepressants and benzodiazepines (sedatives used to treat anxiety and sleeplessness) were involved in about a quarter of such visits (26%). Drugs acting on the central nervous system were involved in 23 percent of the visits, with the most frequent type being opioid/opiate pain medications, which were involved in 14 percent of the total alcohol/drug combination visits.

Table 10
ED visits involving drugs and alcohol taken together, by most frequent combinations, 2007
Drugs reported with alcohol (1) ED visits (2,3) Percent of ED visits (3) Rate of ED visits per 100,000 population (3,4)
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(3) ED visits often involve multiple drugs. Such visits will appear multiple times in this table (e.g., a visit involving both cocaine and marijuana will appear twice in this table). The sum of visits or rates by drug will be greater than the total, and the sum of percentages by drug will be greater than 100.
(4) All rates are ED visits per 100,000 population. Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States. Population estimates are drawn from the 2007 U.S. Census Bureau Postcensal Resident Population National Population Dataset as of July 1, 2007.
(5) Miscellaneous pain medications include acetaminophen, tramadol, and pain medications that were not specified by name. It does not include nonsteroidal anti-inflammatory drugs (such as ibuprofen) or salicylates (such as aspirin).
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, drugs with alcohol 497,283 100.0 163.3
Illicit drugs 307,839   61.9 101.1
Cocaine 194,277   39.1   63.8
Marijuana 117,249   23.6   38.5
Heroin   37,318     7.5   12.3
Stimulants   17,956     3.6     5.9
Pharmaceuticals 259,347   52.2   85.2
Psychotherapeutic agents 131,164   26.4   43.1
Antidepressants   29,116     5.9     9.6
Benzodiazepines   84,849   17.1   27.9
Alprazolam   31,867     6.4   10.5
Clonazepam   14,977     3.0     4.9
Central nervous system agents 114,477   23.0   37.6
Opioid/opiate pain medications   71,108   14.3   23.4
Hydrocodone   21,749     4.4     7.1
Oxycodone   20,734     4.2     6.8
Miscellaneous pain medications (5)   21,313     4.3     7.0
Muscle relaxants   11,323     2.3     3.7

The rate of ED visits per 100,000 population for males (207.3) was higher than that for females (120.5) (Table 11 and Figure 3). Rates by age group showed a general pattern of being lower for those younger than 18 or older than 54, and higher for those aged 18 to 54.

Table 11
ED visits involving drugs and alcohol taken together, by patient demographics, 2007
Patient demographics ED visits (1) Percent of ED visits Rates of ED visits per 100,000 population (2)
(1) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(2) All rates are ED visits per 100,000 population. Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States. Population estimates are drawn from the 2007 U.S. Census Bureau Postcensal Resident Population National Population Dataset as of July 1, 2007.
NOTE: RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed. A dash (—) indicates a blank cell. Rates are not provided for race and ethnicity subgroups because of data limitations.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, drugs with alcohol 497,283 100.0 163.3
Gender         —     —     —
Male 311,221   62.6 207.3
Female 186,010   37.4 120.5
Unknown           *       *       *
Age         —     —     —
0–5 years           *       *       *
6–11 years           *       *       *
12–17 years   26,403     5.3 105.3
18–20 years   32,308     6.5 249.5
21–24 years   49,294     9.9 291.4
25–29 years   60,108   12.1 280.0
30–34 years   51,930   10.4 263.4
35–44 years 135,872   27.3 318.8
45–54 years 105,927   21.3 238.6
55–64 years   25,979     5.2   77.2
65 years and older     9,169     1.8   23.7
Unknown           *       *       *
Race/ethnicity         —     —     —
White 285,865   57.5     —
Black 112,015   22.5     —
Hispanic   53,642   10.8     —
Other or two or more race/ethnicities     5,532     1.1     —
Unknown   40,230     8.1     —

Figure 3
Rates of ED visits per 100,000 population involving alcohol, by age and gender, 2007

Figure 3   D

SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).

Considering race/ethnicity, 57 percent of patients were white, 23 percent were black, 11 percent were Hispanic, 1 percent were of other or multiple race/ethnic groups, and 8 percent were of unknown race/ethnicity. Unfortunately, DAWN is unable to produce population-based rates for race/ethnicity categories. Race/ethnicity information in ED records is often missing or is very limited. By necessity, DAWN uses a simplified set of race/ethnicity categories that is incompatible with the categories used by the U.S. Census Bureau to report population by race/ethnicity. Therefore, the population denominators that would enable DAWN to produce rates are not available.

The disposition of the drug/alcohol combination ED visits (i.e., where the patient went after discharge from the ED) is shown in Table 12. The majority (53%) were treated and released, and slightly more than a quarter (28%) of patients were admitted to the hospital. Of the remaining patients (19%), most were transferred to another health care facility.

Table 12
ED visits involving drugs and alcohol taken together, by patient disposition, 2007
Patient disposition ED visits (1) Percent of ED visits Rates of ED visits per 100,000 population (2)
(1) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(2) All rates are ED visits per 100,000 population. Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States. Population estimates are drawn from the 2007 U.S. Census Bureau Postcensal Resident Population National Population Dataset as of July 1, 2007.
NOTE: RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, drugs with alcohol 497,283 100.0 163.3
Treated and released 262,548   52.8   86.2
Discharged home 210,273   42.3   69.1
Released to police/jail   15,167     3.0     5.0
Referred to detox/treatment   37,108     7.5   12.2
Admitted to this hospital 140,569   28.3   46.2
ICU/critical care   31,524     6.3   10.4
Surgery     1,039     0.2     0.3
Chemical dependency/detox   17,151     3.4     5.6
Psychiatric unit   39,474     7.9   13.0
Other inpatient unit   51,381   10.3   16.9
Other disposition   94,167   18.9   30.9
Transferred   67,907   13.7   22.3
Left against medical advice     8,299     1.7     2.7
Died        572     0.1     0.2
Other           *       *       *
Not documented     4,887     1.0     1.6

Nearly half of all patients received some type of follow-up treatment after their ED visit, whether it was specifically a referral to a drug detox/dependency program, admission to the hospital, or transfer to another health care facility.

ED visits involving alcohol use by patients aged 12 to 17 and 18 to 20

In 2007, for patients aged 12 to 20, almost 59,000 ED visits involved alcohol use with other drugs, and more than 136,000 involved alcohol alone. More than two thirds (70%) of the ED visits involving any type of alcohol use for this age group were associated with alcohol alone and no other drugs. Comparing patients aged 12 to 17 with those aged 18 to 20, the rates for use of alcohol alone were 223.2 and 619.9, respectively (Table 13 and Figure 4), whereas the rates for use of alcohol with other drugs were 105.3 and 249.5, respectively. Within age groups, the rate of ED visits involving the use of alcohol alone is more than double that of alcohol plus drugs, and between age groups, the rate for those aged 18 to 20 is more than double that of younger patients aged 12 to 17.

Table 13
ED visits involving alcohol, by presence of other drugs, by age groups 12 to 17 and 18 to 20, 2007
Alcohol use category (1) ED visits (2) Rate of ED visits per 100,000 population (3) RSE (%) 95% CI:
Lower bound (ED visits)
95% CI:
Upper bound (ED visits)
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(3) All rates are ED visits per 100,000 population. Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States. Population estimates are drawn from the 2007 U.S. Census Bureau Postcensal Resident Population National Population Dataset as of July 1, 2007.
NOTE: CI = confidence interval. RSE = relative standard error.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Alcohol with drugs, patients aged 12 to 17 26,403 105.3 13.0 19,651 33,156
Alcohol with drugs, patients aged 18 to 20 32,308 249.5 10.3 25,773 38,843
Alcohol alone, patients aged 12 to 17 55,960 223.2 10.7 44,214 67,707
Alcohol alone, patients aged 18 to 20 80,255 619.9   8.4 66,992 93,518

Figure 4
Rates of ED visits per 100,000 population involving alcohol in combination and alcohol alone, by age groups 12 to 17 and 18 to 20, 2007

Figure 4   D

SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).

Trends in ED visits involving alcohol, 2004–2007

This section presents the trends in the estimates of ED visits involving alcohol for the period 2004 through 2007 (Table 14). Differences between years are presented in terms of the percentage increase or decrease in visits in 2007 compared with the estimates for the previous 3 years. Only statistically significant changes are discussed and displayed in the table.

Table 14
ED visits involving alcohol, by presence of other drugs, 2004–2007
Alcohol use category (1) ED visits, 2004 (2) ED visits, 2005 (2) ED visits, 2006 (2) ED visits, 2007 (2) Percent change 2004, 2007 (3) Percent change 2005, 2007 (3) Percent change 2006, 2007 (3)
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(3) This column denotes statistically significant (p < 0.05) increases or decreases between estimates for the periods shown.
NOTE: A dash (—) indicates a blank cell.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, alcohol 674,914 527,198 577,521 634,652
Alcohol in combination 523,926 416,599 450,817 497,283
Alcohol alone 150,988 110,599 126,704 137,369

Looking across patients of all ages, no significant changes were found from 2004 to 2007 in the number of ED visits related to drinking alcohol, irrespective of whether other drugs were involved (Table 14).

However, when the data are broken down by age groups, from 2005 to 2007 ED visits involving alcohol in patients aged 12 to 17 rose significantly (32%; Table 15). The 2007 total was more than 80,000 visits (82,364) by patients aged 12 to 17. A large part of that increase reflects the jump of 31 percent in alcohol-alone visits for patients aged 12 to 17. This suggests that alcohol use is starting at an earlier age, with incumbent morbidity as evidenced in the rise in ED visits. NIAAA cautions that persons who start drinking at an early age—for example, at age 14 or younger—are at much higher risk of developing alcohol problems at some point in their lives than those who start drinking at age 21 or after.

Table 15
ED visits involving alcohol, by presence of other drugs, by age groups 12 to 17 and 18 to 20, 2004–2007
Alcohol use category (1) ED visits, 2004 (2) ED visits, 2005 (2) ED visits, 2006 (2) ED visits, 2007 (2) Percent change 2004, 2007 (3) Percent change 2005, 2007 (3) Percent change 2006, 2007 (3)
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(3) This column denotes statistically significant (p < 0.05) increases or decreases between estimates for the periods shown.
NOTE: A dash (—) indicates a blank cell.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, alcohol, aged 12 to 17   67,589 62,459   76,760   82,364  32
Total ED visits, alcohol, aged 18 to 20 135,313 95,166 105,675 112,563
Alcohol with drugs, aged 12 to 17   21,555 19,720   24,418   26,403
Alcohol with drugs, aged 18 to 20   31,926 27,784   31,702   32,308
Alcohol alone, aged 12 to 17   46,034 42,739   52,342   55,960  31
Alcohol alone, aged 18 to 20 103,387 67,382   73,973   80,255


NONMEDICAL USE OF PHARMACEUTICALS

ED visits involving nonmedical use of pharmaceuticals, 2007

As used by DAWN, nonmedical use of pharmaceuticals includes taking more than the prescribed dose of a prescription pharmaceutical or more than the recommended dose of an over-the-counter pharmaceutical or supplement; taking a pharmaceutical prescribed for another individual; deliberate poisoning with a pharmaceutical by another person; and documented misuse or abuse of a prescription drug, an over-the-counter pharmaceutical, or a dietary supplement. Nonmedical use of pharmaceuticals may involve pharmaceuticals alone or pharmaceuticals in combination with illicit drugs or alcohol. DAWN reporters are careful to distinguish appropriate medical use from nonmedical, or inappropriate, use, and only the latter is included in this grouping.

DAWN tries to capture only pharmaceuticals that are related to the ED visit and actively discourages reporting of current medications that are unrelated to the visit. Given the limitations of medical record documentation, though, it is not always possible to distinguish and exclude current medications that are unrelated to the visit. This limitation may have the effect of overstating the variety of pharmaceuticals involved in ED visits.

For 2007, DAWN estimates that 855,838 (CI: 719,765 to 991,910) ED visits involved nonmedical use of prescription medicines, over-the-counter drugs, or other type of pharmaceutical (Table 16). Multiple drug involvement occurred in 52 percent of visits, and alcohol was involved in 19 percent.

Table 16
ED visits involving nonmedical use of pharmaceuticals, 2007
Drug category and selected drugs (1) ED visits (2,3) Percent of ED visits (3) RSE (%) 95% CI:
Lower bound
95% CI:
Upper bound
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(3) ED visits often involve multiple drugs. Such visits will appear multiple times in this table (e.g., a visit involving both acetaminophen and morphine will appear twice in this table). The sum of visits or rates by drug will be greater than the total, and the sum of percentages by drug will be greater than 100.
NOTE: CI = confidence interval. CNS = central nervous system. NOS = not otherwise specified. NTA = not tabulated above. RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, nonmedical use 855,838 100.0   8.1 719,765 991,910
Visits involving a single drug 415,196   48.5 11.8 318,906 511,485
Visits involving multiple drugs 440,642   51.5   6.9 381,404 499,880
Visits involving alcohol 162,072   18.9   8.6 134,625 189,519
PSYCHOTHERAPEUTIC AGENTS 353,931   41.4   6.4 309,284 398,579
Antidepressants   82,009     9.6   6.9   70,856   93,163
MAO inhibitors            *        *      *            *            *
SSRI antidepressants   37,446     4.4   9.6   30,397   44,496
Tricyclic antidepressants   16,600     1.9 16.7   11,182   22,019
Miscellaneous antidepressants     9,687     1.1 13.7     7,076   12,297
Antipsychotics   52,752     6.2   9.9   42,522   62,982
Anxiolytics, sedatives, and hypnotics 259,983   30.4   8.0 219,443 300,523
Barbiturates     9,877     1.2 19.8     6,044   13,710
Benzodiazepines 218,640   25.5   9.1 179,649 257,632
Alprazolam   80,313     9.4 12.2   61,101   99,525
Clonazepam   40,920     4.8 10.8   32,249   49,591
Diazepam   19,674     2.3 11.9   15,103   24,245
Lorazepam   26,213     3.1 12.1   20,006   32,419
Benzodiazepines NOS   55,346     6.5 23.9   29,419   81,274
Misc. anxiolytics, sedatives, and hypnotics   43,960     5.1 10.8   34,660   53,260
Diphenhydramine   12,539     1.5 13.1     9,307   15,770
Hydroxyzine     2,447     0.3 21.5     1,417     3,478
Zolpidem   18,464     2.2 13.0   13,756   23,173
Anxiolytics, sedatives, and hypnotics NOS     3,364     0.4 18.6     2,135     4,593
CNS stimulants   18,561     2.2   9.6   15,069   22,052
Amphetamine-dextroamphetamine     6,372     0.7 16.7     4,292     8,452
Caffeine     2,165     0.3 27.4     1,002     3,329
Dextroamphetamine            *        *      *            *            *
Methylphenidate     4,782     0.6 19.8     2,925     6,640
CENTRAL NERVOUS SYSTEM AGENTS 415,354   48.5   7.6 353,459 477,248
Pain medications 363,621   42.5   8.0 306,503 420,740
Antimigraine agents     2,284     0.3 28.9        991     3,577
Cox-2 inhibitors        635     0.1 41.3        121     1,148
Opiates/opioids 286,521   33.5   9.2 235,089 337,954
Opiates/opioids, unspecified   52,997     6.2 12.6   39,943   66,050
Narcotic pain medications 237,143   27.7 10.4 188,610 285,676
Buprenorphine/combinations     7,136     0.8 31.1     2,786   11,486
Codeine/combinations     5,648     0.7 16.1     3,862     7,433
Fentanyl/combinations   15,947     1.9 22.0     9,080   22,813
Hydrocodone/combinations   65,734     7.7 13.3   48,584   82,883
Hydromorphone/combinations     9,497     1.1 25.7     4,709   14,285
Meperidine/combinations        997     0.1 32.2        369     1,626
Methadone   53,950     6.3 14.8   38,278   69,621
Morphine/combinations   29,591     3.5 37.0     8,118   51,065
Oxycodone/combinations   76,587     8.9 12.4   58,015   95,160
Propoxyphene/combinations     7,401     0.9 16.9     4,946     9,856
Nonsteroidal anti-inflammatory agents   30,822     3.6 10.4   24,568   37,076
Ibuprofen   20,892     2.4 10.1   16,752   25,032
Naproxen     7,208     0.8 17.8     4,695     9,720
Salicylates/combinations     9,724     1.1 12.9     7,269   12,178
Miscellaneous pain medications/combinations   56,534     6.6 10.7   44,645   68,424
Acetaminophen/combinations   43,872     5.1 12.9   32,801   54,942
Tramadol/combinations     8,039     0.9 20.0     4,895   11,184
Tramadol     7,662     0.9 20.1     4,647   10,678
Acetaminophen-tramadol            *        *      *            *            *
Pain medication combinations NTA     2,120     0.2 21.5     1,228     3,013
Anorexiants        758     0.1 30.7        302     1,213
Anticonvulsants   35,403     4.1   8.8   29,298   41,507
Antiemetic/antivertigo agents     1,646     0.2 30.8        651     2,640
Anti-Parkinson agents     3,764     0.4 18.2     2,425     5,104
General anesthetics            *        *      *            *            *
Muscle relaxants   40,769     4.8 19.3   25,371   56,166
Carisoprodol   27,128     3.2 27.3   12,600   41,656
Cyclobenzaprine     6,197     0.7 16.2     4,228     8,166
Miscellaneous CNS agents        924     0.1 40.2        195     1,652
RESPIRATORY AGENTS   31,008     3.6 12.4   23,469   38,547
Antihistamines     5,096     0.6 18.2     3,273     6,918
Bronchodilators     3,043     0.4 20.0     1,849     4,238
Decongestants     1,758     0.2 35.8        525     2,991
Expectorants     2,293     0.3 19.7     1,406     3,180
Upper respiratory combinations   16,677     1.9 13.7   12,200   21,155
Respiratory agents NTA     4,655     0.5 19.3     2,897     6,414
CARDIOVASCULAR AGENTS   35,608     4.2 11.2   27,807   43,408
Antiadrenergic agents, centrally acting     4,751     0.6 16.5     3,213     6,289
Beta-adrenergic blocking agents   11,668     1.4 16.8     7,831   15,505
Calcium channel blocking agents     4,493     0.5 18.9     2,833     6,154
Diuretics     5,467     0.6 23.1     2,988     7,946
Cardiovascular agents NTA   17,879     2.1 13.0   13,319   22,440

At 49 percent, central nervous system agents were the most common type of drugs reported in the nonmedical-use category of ED visits. Specific drugs seen at high levels were methadone and single-ingredient and combination forms of oxycodone and hydrocodone, all narcotic painkillers.11 Once the margin of error is taken into account, ED visits involving nonmedical use of oxycodone, hydrocodone, and methadone appeared at similar levels:

The nonopioid pain medication acetaminophen and muscle relaxants each showed up as being involved in about 5 percent of nonmedical-use visits. The most common muscle relaxant was carisoprodol (3%). Anticonvulsants and nonsteroidal anti-inflammatory agents (e.g., ibuprofen, naproxen) each showed up in about 4 percent of visits.

At 41 percent, psychotherapeutic agents were the second most frequent type of drug reported in the nonmedical-use category of ED visits. Anxiolytics accounted for the largest part of that, followed by antidepressants and antipsychotics (e.g., quetiapine).

Benzodiazepines, an anxiolytic prescribed for a wide range of conditions but predominately anxiety and insomnia, were involved in 26 percent of all ED visits related to nonmedical use. DAWN estimates that 218,640 (CI: 179,649 to 257,632) ED visits were associated with nonmedical use of benzodiazepines. The specific types of benzodiazepines found were

Benzodiazepines, without a specific ingredient named, appeared in an additional 55,346 (CI: 29,419 to 81,274) ED visits, or 6 percent.

Other types of anxiolytics, sedatives, and hypnotics found at notable levels included

The two other major categories of pharmaceuticals are respiratory agents and cardiovascular agents, and each was involved in about 4 percent of nonmedical-use ED visits.

Of the total number of ED visits involving the nonmedical use of pharmaceuticals, the majority of these visits (52%) involved multiple drugs. Specifically,

When population size and the margin of error are taken into account, visits for nonmedical use of pharmaceuticals did not differ between females (286 visits per 100,000 population) and males (275 visits per 100,000 population; Table 17 and Figure 5). ED visit rates for patients aged 18 to 44 were all more than 400 visits per 100,000 population.

Table 17
ED visits and rates involving nonmedical use of pharmaceuticals, by patient demographics, 2007
Patient demographics ED visits (1) Percent of ED visits Rate of ED visits per 100,000 population (2)
(1) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(2) All rates are ED visits per 100,000 population. Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States. Population estimates are drawn from the 2007 U.S. Census Bureau Postcensal Resident Population National Population Dataset as of July 1, 2007.
NOTE: RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed. A dash (—) indicates a blank cell. Rates are not provided for race and ethnicity subgroups because of data limitations.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, nonmedical use 855,838 100.0 281.1
Gender         —      —      —
Male 413,581   48.3 275.4
Female 441,763   51.6 286.3
Unknown            *        *        *
Age         —      —      —
0–5 years     4,986     0.6   19.9
6–11 years     4,450     0.5   18.6
12–17 years   68,586     8.0 273.6
18–20 years   57,558     6.7 444.6
21–24 years   77,132     9.0 456.0
25–29 years   95,234   11.1 443.7
30–34 years   81,131     9.5 411.6
35–44 years 172,012   20.1 403.6
45–54 years 162,884   19.0 367.0
55–64 years   69,085     8.1 205.2
65 years and older   62,670     7.3 161.6
Unknown            *        *        *
Race/ethnicity         —      —      —
White 599,668   70.1      —
Black 105,901   12.4      —
Hispanic   74,789     8.7      —
Other or two or more race/ethnicities   12,043     1.4      —
Unknown   63,436     7.4      —

Figure 5
Rates of ED visits per 100,000 population involving nonmedical use of pharmaceuticals, by age and gender, 2007

Figure 5   D

SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).

In terms of race and ethnicity, 70 percent of visits related to nonmedical use of pharmaceuticals involved patients who were white. Unfortunately, DAWN is unable to produce population-based rates for race/ethnicity categories. Race/ethnicity information in ED records is often missing or is very limited. By necessity, DAWN uses a simplified set of race/ethnicity categories that is incompatible with the categories used by the U.S. Census Bureau to report population by race/ethnicity. Therefore, the population denominators that would enable DAWN to produce rates are not available.

Patient disposition after ED visits associated with nonmedical use of pharmaceuticals appears in Table 18. The majority of patients (59%) were treated and released, more than a quarter (27%) were admitted to the hospital, and the balance (14%) had other outcomes.

Table 18
ED visits and rates involving nonmedical use of pharmaceuticals, by patient disposition, 2007
Patient disposition ED visits (1) Percent of ED visits Rate of ED visits per 100,000 population (2)
(1) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(2) All rates are ED visits per 100,000 population. Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States. Population estimates are drawn from the 2007 U.S. Census Bureau Postcensal Resident Population National Population Dataset as of July 1, 2007.
NOTE: RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, nonmedical use 855,838 100.0 281.1
Treated and released 502,761   58.7 165.1
Discharged home 456,385   53.3 149.9
Released to police/jail   16,791     2.0     5.5
Referred to detox/treatment   29,585     3.5     9.7
Admitted to this hospital 233,300   27.3   76.6
ICU/critical care   71,730     8.4   23.6
Surgery           *     0.2        *
Chemical dependency/detox     3,064     0.4     1.0
Psychiatric unit   50,013     5.8   16.4
Other inpatient unit 106,880   12.5   35.1
Other follow-up 119,777   14.0   39.3
Transferred   83,547     9.8   27.4
Left against medical advice   18,559     2.2     6.1
Died     1,379     0.2     0.5
Other   10,584     1.2     3.5
Not documented     5,707     0.7     1.9

Trends in ED visits involving nonmedical use of pharmaceuticals, 2004–2007

This section presents the trends in the estimates of ED visits involving nonmedical use of pharmaceuticals for the period 2004 through 2007 (Table 19). Differences between years are presented in terms of the percentage increase or decrease in visits in 2007 compared with the estimates for the previous 3 years. Only statistically significant changes are discussed and displayed in the table.

Table 19
Trends in ED visits involving nonmedical use of pharmaceuticals, by selected drugs, 2004–2007
Drug category and selected drugs (1) ED visits, 2004 (2,3) ED visits, 2005 (2,3) ED visits, 2006 (2,3) ED visits, 2007 (2,3) Percent change
2004, 2007 (4)
Percent change 2005, 2007 (4) Percent change 2006, 2007 (4)
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(3) ED visits often involve multiple drugs. Such visits will appear multiple times in this table (e.g., a visit involving both acetaminophen and morphine will appear twice in this table). The sum of visits or rates by drug will be greater than the total, and the sum of percentages by drug will be greater than 100.
(4) This column denotes statistically significant (p < 0.05) increases or decreases between estimates for the periods shown.
NOTE: CNS = central nervous system. NOS = not otherwise specified. NTA = not tabulated above. RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed. A dash (—) indicates a blank cell.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, nonmedical use 536,247 669,214 741,425 855,838   60  —   15
PSYCHOTHERAPEUTIC AGENTS 247,324 308,655 323,999 353,931   43  —  —
Antidepressants   66,917   67,051   79,682   82,009  —  —  —
MAO inhibitors           *           *           *           *  —  —  —
SSRI antidepressants   32,285   30,374   35,370   37,446  —  —  —
Tricyclic antidepressants   12,412   14,515   16,564   16,600  —  —  —
Miscellaneous antidepressants     9,414     7,452     7,561     9,687  —  —  —
Antipsychotics   35,198   44,393   44,733   52,752   50  —  —
Anxiolytics, sedatives, and hypnotics 177,394 227,486 233,875 259,983   47  —  —
Barbiturates   11,721   14,693   10,991     9,877  —  -33  —
Benzodiazepines 143,546 189,704 195,625 218,640   52  —  —
Alprazolam   46,526   57,419   65,236   80,313   73   40   23
Clonazepam   28,178   30,648   33,557   40,920   45   34  —
Diazepam   15,619   18,433   19,936   19,674  —  —  —
Lorazepam   17,674   23,210   23,720   26,213   48  —  —
Benzodiazepines NOS   36,039   61,486   58,347   55,346  —  —  —
Misc. anxiolytics, sedatives, and hypnotics   31,554   35,561   40,626   43,960   39  —  —
Diphenhydramine   10,452   10,294   12,291   12,539  —  —  —
Hydroxyzine     2,363     2,179     2,678     2,447  —  —  —
Zolpidem   12,792   14,730   17,257   18,464   44  —  —
Anxiolytics, sedatives, and hypnotics NOS     2,657     4,421     3,629     3,364  —  —  —
CNS stimulants     9,801   10,965   13,892   18,561   89   69  —
Amphetamine-dextroamphetamine     2,303     2,669     5,027     6,372 177 139  —
Caffeine     2,736     4,567     4,407     2,165  —  -53  -51
Dextroamphetamine           *           *           *           *  —  —  —
Methylphenidate     2,446     2,519     2,192     4,782  —   90 118
CENTRAL NERVOUS SYSTEM AGENTS 282,296 336,900 373,138 415,354   47   23  —
Pain medications 241,578 294,251 323,579 363,621   51   24  —
Antimigraine agents        868     1,018     1,191     2,284  — 124  —
Cox-2 inhibitors     1,935        765           *        635  —  —  —
Opiates/opioids 172,726 217,594 247,669 286,521   66   32   16
Opiates/opioids, unspecified   31,846   52,670   50,978   52,997   66  —  —
Narcotic pain medications 144,644 168,376 201,280 237,143   64   41  —
Buprenorphine/combinations           *           *     4,440     7,136  — 224  —
Codeine/combinations     7,171     6,180     6,928     5,648  —  —  —
Fentanyl/combinations     9,823   11,211   16,012   15,947   62   42  —
Hydrocodone/combinations   39,844   47,192   57,550   65,734   65   39  —
Hydromorphone/combinations     3,385     4,714     6,780     9,497 181 101  —
Meperidine/combinations        782        383     1,440        997  —  —  —
Methadone   36,806   42,684   45,130   53,950   47  —  —
Morphine/combinations   13,966   15,762   20,416   29,591  —  —  —
Oxycodone/combinations   41,701   52,943   64,888   76,587   84   45  —
Propoxyphene/combinations     6,744     7,648     6,220     7,401  —  —  —
Nonsteroidal anti-inflammatory agents   27,362   28,837   27,662   30,822  —  —  —
Ibuprofen   22,127   22,268   20,541   20,892  —  —  —
Naproxen     4,715     5,190     6,651     7,208  —  —  —
Salicylates/combinations     9,580   12,123   10,399     9,724  —  —  —
Miscellaneous pain medications/combinations   44,857   51,881   54,313   56,534  —  —  —
Acetaminophen/combinations   39,167   43,558   44,314   43,872  —  —  —
Tramadol/combinations     4,849     5,918     6,048     8,039  —  —  —
Tramadol     3,948     5,427     5,961     7,662  —  —  —
Acetaminophen-tramadol        909           *           *           *  —  —  —
Pain medication combinations NTA        977        653        898     2,120 117 225  —
Anorexiants           *     1,757     1,168        758  —  -57  —
Anticonvulsants   28,652   27,641   31,169   35,403  —  —  —
Antiemetic/antivertigo agents     1,678     1,771     1,360     1,646  —  —  —
Anti-Parkinson agents     2,472     1,692     3,816     3,764  — 123  —
General anesthetics           *           *           *           *  —  —  —
Muscle relaxants   25,934   33,695   38,918   40,769  —  —  —
Carisoprodol   14,736   20,082   24,505   27,128  —  —  —
Cyclobenzaprine     6,183     7,629     7,142     6,197  —  —  —
Miscellaneous CNS agents        869        900        999        924  —  —  —
RESPIRATORY AGENTS   22,286   28,017   28,867   31,008  —  —  —
Antihistamines     5,761     4,429     4,130     5,096  —  —  —
Bronchodilators     2,294     3,043     2,920     3,043  —  —  —
Decongestants     1,864     1,309     1,511     1,758  —  —  —
Expectorants        832     1,960     2,125     2,293 176  —  —
Upper respiratory combinations   10,314   15,837   15,115   16,677   62  —  —
Respiratory agents NTA     2,903     3,692     4,296     4,655  —  —  —
CARDIOVASCULAR AGENTS   27,396   37,095   36,343   35,608  —  —  —
Antiadrenergic agents, centrally acting     3,616     5,125     4,810     4,751  —  —  —
Beta-adrenergic blocking agents     7,094     9,824   11,729   11,668  —  —  —
Calcium channel blocking agents     3,115     5,434     5,227     4,493  —  —  —
Diuretics     3,625     5,332     5,102     5,467  —  —  —
Cardiovascular agents NTA   14,930   18,881   17,338   17,879  —  —  —


ED visits related to nonmedical use of pharmaceuticals increased 60 percent in the period from 2004 to 2007. Among the drugs most frequently implicated in nonmedical use, the following changes from 2004 to 2007 are notable:

DRUG-RELATED SUICIDE ATTEMPTS

ED visits involving drug-related suicide attempts, 2007

DAWN collects information on ED visits after suicide attempts that involve drugs. These attempts are not limited to drug overdoses. Suicide attempts involving firearms, for example, are included as DAWN cases if drugs were involved at all at the time of the suicide attempt.13

DAWN estimates 197,053 (CI: 164,564 to 229,542) ED visits for drug-related suicide attempts in 2007 (Table 20). Almost all (94%) involved a prescription drug or over-the-counter medication. More than half (57%) involved psychotherapeutic agents, nearly half (48%) involved central nervous system agents, just under a third (29%) involved alcohol, and about a fifth (19%) involved illicit drugs.14 Nearly two thirds (63%) of ED visits for drug-related suicide attempts involved multiple drugs.

Table 20
ED visits involving drug-related suicide attempts, by selected drugs, 2007
Drug category and selected drugs (1) ED visits (2,3) Percent of ED visits (3) RSE (%) 95% CI: Lower bound 95% CI: Upper bound
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(3) ED visits often involve multiple drugs. Such visits will appear multiple times in this table (e.g., a visit involving both cocaine and marijuana will appear twice in this table). The sum of visits or rates by drug will be greater than the total, and the sum of percentages by drug will be greater than 100.
NOTE: CI = confidence interval. CNS = central nervous system. NOS = not otherwise specified. NTA = not tabulated above. RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, suicide attempts 197,053 100.0   8.4 164,564 229,542
Visits involving a single drug   73,346   37.2   8.9   60,507   86,184
Visits involving multiple drugs 123,707   62.8 10.0   99,479 147,935
Visits involving illicit drugs   37,319   18.9 24.0   19,780   54,859
Visits involving alcohol   57,319   29.1   9.7   46,439   68,199
Visits involving pharmaceuticals 185,307   94.0   8.3 155,213 215,400
Alcohol   57,319   29.1   9.7   46,439   68,199
Alcohol in combination   56,434   28.6 10.0   45,334   67,533
Alcohol alone           *       * 58.6           *           *
Non-alcohol illicit   37,319   18.9 24.0   19,780   54,859
Cocaine   26,462   13.4 30.0   10,901   42,022
Heroin     4,444     2.3 33.2     1,556     7,332
Marijuana   12,115     6.1 24.1     6,404   17,826
Stimulants     2,665     1.4 29.4     1,130     4,201
Amphetamines        878     0.4 29.3        374     1,381
Methamphetamine     1,795     0.9 37.9        461     3,130
MDMA (Ecstasy)        481     0.2 47.7          32        931
GHB           *       *     *           *           *
Flunitrazepam (Rohypnol)           *       *     *           *           *
Ketamine           *       *     *           *           *
LSD           *       *     *           *           *
PCP        768     0.4 40.2        163     1,374
Miscellaneous hallucinogens           *       *     *           *           *
Inhalants           *       *     *           *           *
Combinations NTA           *       *     *           *           *
PSYCHOTHERAPEUTIC AGENTS 112,768   57.2   8.2   94,681 130,854
Antidepressants   38,870   19.7 11.1   30,418   47,322
MAO inhibitors           *       *     *           *           *
SSRI antidepressants   18,884     9.6 10.6   14,949   22,820
Tricyclic antidepressants     4,152     2.1 33.7     1,408     6,896
Miscellaneous antidepressants     3,939     2.0 17.2     2,610     5,267
Antipsychotics   25,479   12.9 18.5   16,239   34,720
Anxiolytics, sedatives, and hypnotics   72,637   36.9   6.7   63,037   82,236
Barbiturates     1,663     0.8 39.3        382     2,945
Benzodiazepines   53,509   27.2   7.3   45,860   61,157
Alprazolam   19,167     9.7 11.3   14,908   23,425
Clonazepam   14,455     7.3   9.4   11,787   17,122
Diazepam     6,912     3.5 19.4     4,287     9,537
Lorazepam     9,527     4.8 15.0     6,730   12,325
Benzodiazepines NOS     4,594     2.3 35.4     1,407     7,781
Misc. anxiolytics, sedatives, and hypnotics   23,349   11.8   8.8   19,329   27,369
Diphenhydramine     7,618     3.9 16.1     5,221   10,015
Hydroxyzine     2,027     1.0 32.1        751     3,302
Zolpidem     7,403     3.8 14.2     5,349     9,458
Anxiolytics, sedatives, and hypnotics NOS     2,274     1.2 28.2     1,017     3,531
CNS stimulants     2,208     1.1 23.2     1,203     3,212
Amphetamine-dextroamphetamine        576     0.3 33.6        196        955
Caffeine           *       *     *           *           *
Dextroamphetamine           *       *     *           *           *
Methylphenidate     1,002     0.5 41.3        190     1,815
CNS AGENTS   94,644   48.0   8.9   78,077 111,210
Pain medications   78,948   40.1 10.3   62,958   94,938
Antimigraine agents           *       *     *           *           *
Cox-2 inhibitors           *       *     *           *           *
Opiates/opioids   31,476   16.0 10.4   25,078   37,874
Opiates/opioids, unspecified     1,893     1.0 23.2     1,032     2,754
Narcotic pain medications   29,886   15.2 10.1   23,945   35,827
Buprenorphine/combinations           *       *     *           *           *
Codeine/combinations     1,637     0.8 28.0        737     2,536
Fentanyl/combinations           *       * 54.4           *           *
Hydrocodone/combinations   13,238     6.7 12.5     9,989   16,488
Hydromorphone/combinations        796     0.4 26.1        389     1,202
Meperidine/combinations           *       *     *           *           *
Methadone     3,192     1.6 26.0     1,566     4,817
Morphine/combinations     1,690     0.9 41.9        303     3,076
Oxycodone/combinations     9,351     4.7 23.9     4,972   13,731
Propoxyphene/combinations     1,754     0.9 22.1        994     2,514
Ibuprofen   14,057     7.1 14.7   10,007   18,107
Salicylates/combinations     5,976     3.0 17.3     3,953     7,998
Miscellaneous pain medications/combinations   32,968   16.7 11.7   25,426   40,510
Acetaminophen/combinations   29,861   15.2 12.1   22,783   36,940
Tramadol     2,669     1.4 23.1     1,458     3,880
Acetaminophen-tramadol           *       *     *           *           *
Pain medication combinations NTA     1,147     0.6 25.7        569     1,725
Anorexiants           *       *     *           *           *
Anticonvulsants   11,803     6.0   8.7     9,799   13,806
Antiemetic/antivertigo agents        343     0.2 41.1          66        620
Antiparkinson agents        755     0.4 38.3        188     1,321
General anesthetics           *       *     *           *           *
Muscle relaxants     9,772     5.0 13.7     7,141   12,403
Carisoprodol     4,301     2.2 23.5     2,319     6,282
Cyclobenzaprine     3,839     1.9 22.2     2,169     5,510
Miscellaneous CNS agents           *       *     *           *           *
RESPIRATORY AGENTS   10,175     5.2 16.4     6,903   13,448
Antihistamines     3,813     1.9 23.5     2,058     5,568
Bronchodilators           *       *     *           *           *
Decongestants        805     0.4 49.0          32     1,578
Expectorants        649     0.3 36.2        188     1,110
Upper respiratory combinations     4,067     2.1 18.1     2,621     5,512
Respiratory agents NTA     1,114     0.6 24.5        579     1,648
CARDIOVASCULAR AGENTS     7,873     4.0 14.8     5,591   10,154
Antiadrenergic agents, centrally acting        790     0.4 27.2        368     1,211
Beta-adrenergic blocking agents     2,501     1.3 21.5     1,449     3,554
Calcium channel blocking agents        601     0.3 42.8          96     1,106
Diuretics     1,360     0.7 26.0        668     2,052
Cardiovascular agents NTA     4,344     2.2 21.0     2,559     6,128

The more common drugs involved in suicide attempts include the following:

After population size and the margin of error are taken into account, the rate of drug-related suicide attempt visits for females (78 visits per 100,000 population) was higher than that for males (52 per 100,000; Table 21 and Figure 6). In respect to age, rates ranged from a low of 10 visits per 100,000 persons for those aged 65 or older to 152 visits for those aged 18 to 20. In general, the rates for patients aged 18 to 44 are higher than the rates for younger and older age groups. In terms of race/ethnicity, 63 percent of the suicide attempts involved patients who were white. Unfortunately, DAWN is unable to produce population-based rates for race/ethnicity categories. Race/ethnicity information in ED records is often missing or is very limited. By necessity, DAWN uses a simplified set of race/ethnicity categories that is incompatible with the categories used by the U.S. Census Bureau to report population by race/ethnicity. Therefore, the population denominators that would enable DAWN to produce rates are not available.

Table 21
ED visits involving drug-related suicide attempts, by patient demographics, 2007
Patient demographics ED visits (1) Percent of ED visits Rates of ED visits per 100,000 population (2)
(1) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(2) All rates are ED visits per 100,000 population. Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States. Population estimates are drawn from the 2007 U.S. Census Bureau Postcensal Resident Population National Population Dataset as of July 1, 2007.
NOTE: RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed. A dash (—) indicates a blank cell. Rates are not provided for race and ethnicity subgroups because of data limitations.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, suicide attempts 197,053 100.0   64.7
Gender         —     —     —
Male   77,299   39.2   51.5
Female 119,745   60.8   77.6
Unknown           *       *       *
Age         —     —     —
0–5 years           *       *       *
6–11 years           *       *       *
12–17 years   21,093   10.7   84.1
18–20 years   19,706   10.0 152.2
21–24 years   18,384     9.3 108.7
25–29 years   26,041   13.2 121.3
30–34 years   20,780   10.5 105.4
35–44 years   45,254   23.0 106.2
45–54 years   31,765   16.1   71.6
55–64 years   10,166     5.2   30.2
65 years and older     3,826     1.9     9.9
Unknown           *       *       *
Race/ethnicity         —     —     —
White 123,762   62.8     —
Black           *       *     —
Hispanic   18,182     9.2     —
Other or two or more race/ethnicities     3,773     1.9     —
Unknown   14,882     7.6     —

Figure 6
Rates of ED visits per 100,000 population involving drug-related suicide attempts, by age and gender, 2007

Figure 6   D

SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).

About half (51%) of the patients attempting suicide were admitted for inpatient hospital care (Table 22). A fifth (20%) were admitted to an ICU/critical care unit; others were admitted to psychiatric units (16%) or other inpatient units (15%). Another 27 percent were transferred to another health care facility; only 13 percent were discharged home. Very few died in the ED. However, DAWN does not record deaths for patients who died before arriving at the ED or patients who died after admission to inpatient units of the hospital.

Table 22
ED visits involving drug-related suicide attempts, by patient disposition, 2007
Patient disposition ED visits (1) Percent of ED visits Rates of ED visits per 100,000 population (2)
(1) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(2) All rates are ED visits per 100,000 population. Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States. Population estimates are drawn from the 2007 U.S. Census Bureau Postcensal Resident Population National Population Dataset as of July 1, 2007.
NOTE: RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update)
Total ED visits, suicide attempts 197,053 100.0 64.7
Treated and released   34,714   17.6 11.4
Discharged home   26,049   13.2   8.6
Released to police/jail     3,401     1.7   1.1
Referred to detox/treatment     5,263     2.7   1.7
Admitted to this hospital 101,060   51.3 33.2
ICU/critical care   39,348   20.0 12.9
Surgery           *        *      *
Chemical dependency/detox        296     0.2   0.1
Psychiatric unit   31,045   15.8 10.2
Other inpatient unit   28,814   14.6   9.5
Other disposition   61,279   31.1 20.1
Transferred   53,121   27.0 17.4
Left against medical advice        337     0.2   0.1
Died           *        *      *
Other           *        *      *
Not documented           *        *      *

Trends in ED visits involving drug-related suicide attempts, 2004–2007

This section presents the trends in the estimates of drug-related ED visits involving suicide attempts for the period 2004 through 2007 (Table 23 and Table 24). Differences between years are presented in terms of the percentage increase or decrease in visits in 2007 compared with the estimates for the previous 3 years. Only statistically significant changes are discussed and displayed in the tables.

ED visits for drug-related suicide attempts were stable from 2004 to 2005, followed by an increase of 30 percent from 2005 to 2007. This increase brought the total number of drug-related ED visits involving suicide attempts to almost 200,000 (197,053 visits) in 2007. The rise in drug-related suicide-attempt ED visits appears to be associated with increases in visits related to psychotherapeutic and central nervous system agents. From 2005 to 2007, ED visits involving psychotherapeutic agents rose by about 30,000 visits, and visits involving central nervous system agents rose by about 28,000 visits.

Specific types of drugs appearing frequently in suicide attempts are listed in Table 23.

Table 23
Drug categories and drugs appearing more frequently in suicide attempts, 2005–2007
Drug category and selected drugs (1) Increase in visits,
2005 to 2007 (2)
Percent increase in visits,
2005 to 2007 (3)
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(3) This column denotes statistically significant (p < 0.05) increases or decreases between estimates for the periods shown.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Benzodiazepines 17,833   50
Clonazepam   5,391   59
Lorazepam   4,345   84
Antidepressants 11,784   44
Opiates/opioid pain medications 11,117   55
Hydrocodone   6,203   88
Oxycodone   5,122 121
Muscle relaxants   3,987   69
Carisoprodol   2,263 111
Antihistamines   2,163 131
Miscellaneous pain medications 10,276   45
Tramadol   1,590 147
Table 24
Trends in ED visits for drug-related suicide attempts, by selected drugs, 2004–2007
Drug category and selected drugs (1) ED visits, 2004 (2,3) ED visits, 2005 (2,3) ED visits, 2006 (2,3) ED visits, 2007 (2,3) Percent change
2004, 2007 (4)
Percent change
2005, 2007 (4)
Percent change
2006, 2007 (4)
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(3) ED visits often involve multiple drugs. Such visits will appear multiple times in this table (e.g., a visit involving both cocaine and marijuana will appear twice in this table). The sum of visits or rates by drug will be greater than the total, and the sum of percentages by drug will be greater than 100.
(4) This column denotes statistically significant (p < 0.05) increases or decreases between estimates for the periods shown.
NOTE: CNS = central nervous system. NOS = not otherwise specified. NTA = not tabulated above. RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed. A dash (—) indicates a blank cell.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, suicide attempts 161,586 151,568 182,805 197,053  —   30  —
Alcohol   48,726   47,891   54,820   57,319  —  —  —
Alcohol in combination   48,080   46,806   54,337   56,434  —  —  —
Alcohol alone        646     1,085        483           *  —  —  —
Non-alcohol illicits   34,763   33,784   42,148   37,319  —  —  —
Cocaine   19,520   19,628   26,510   26,462  —  —  —
Heroin     4,579     3,167     4,265     4,444  —  —  —
Marijuana   12,074   11,955   15,272   12,115  —  —  —
Stimulants     4,535     5,410     4,829     2,665  —  —  -45
Amphetamines     1,560     1,646     2,228        878  —  —  -61
Methamphetamine     3,136     3,853     2,877     1,795  —  —  —
MDMA (Ecstasy)           *        529     1,239        481  —  —  —
GHB           *           *           *           *  —  —  —
Flunitrazepam (Rohypnol)           *           *           *           *  —  —  —
Ketamine           *           *           *           *  —  —  —
LSD           *           *           *           *  —  —  —
PCP           *           *           *        768  —  —  —
Miscellaneous hallucinogens           *           *           *           *  —  —  —
Inhalants           *        794           *           *  —  —  —
Combinations NTA           *           *           *           *  —  —  —
PSYCHOTHERAPEUTIC AGENTS   88,034   82,144 106,128 112,768   28   37  —
Antidepressants   33,366   27,086   36,677   38,870  —   44  —
MAO inhibitors           *           *           *           *  —  —  —
SSRI antidepressants   18,513   13,377   16,973   18,884  —  —  —
Tricyclic antidepressants     3,555     3,008     4,681     4,152  —  —  —
Miscellaneous antidepressants     3,337     2,681     3,806     3,939  —  —  —
Antipsychotics   17,807   17,129   22,491   25,479  —  —  —
Anxiolytics, sedatives, and hypnotics   52,653   52,022   68,177   72,637   38   40  —
Barbiturates     1,949     1,219     2,031     1,663  —  —  —
Benzodiazepines   36,995   35,676   50,431   53,509   45   50  —
Alprazolam   11,354   14,530   15,633   19,167   69  —  —
Clonazepam     9,403     9,064   14,173   14,455  —   59  —
Diazepam     4,630     3,968     5,909     6,912  —  —  —
Lorazepam     6,065     5,182     6,682     9,527  —   84  —
Benzodiazepines NOS     4,426     3,343     7,080     4,594  —  —  —
Misc. anxiolytics, sedatives, and hypnotics   16,790   17,522   21,527   23,349  —  —  —
Diphenhydramine     7,458     6,583     7,756     7,618  —  —  —
Hydroxyzine     2,346     1,795     1,956     2,027  —  —  —
Zolpidem     4,355     4,972     6,674     7,403   70  —  —
Anxiolytics, sedatives, and hypnotics NOS     1,859     2,147     1,406     2,274  —  —  —
CNS stimulants     1,654     1,782     1,949     2,208  —  —  —
Amphetamine-dextroamphetamine           *           *        559        576  —  —  —
Caffeine           *        450           *           *  —  —  —
Dextroamphetamine           *           *           *           *  —  —  —
Methylphenidate           *        818        633     1,002  —  —  —
CENTRAL NERVOUS SYSTEM AGENTS   73,949   66,321   82,442   94,644  —   43  —
Pain medications   61,095   54,858   67,623   78,948  —   44  —
Antimigraine agents           *           *           *           *  —  —  —
Cox-2 inhibitors        807        514           *           *  —  —  —
Opiates/opioids   18,939   20,359   27,185   31,476   66   55  —
Opiates/opioids, unspecified     2,363     2,819     3,129     1,893  —  —  —
Narcotic pain medications   16,928   17,801   24,470   29,886   77   68  —
Buprenorphine/combinations           *           *           *           *  —  —  —
Codeine/combinations     1,750     2,656     2,349     1,637  —  —  —
Fentanyl/combinations           *           *           *           *  —  —  —
Hydrocodone/combinations     7,034     7,035     8,998   13,238   88   88   47
Hydromorphone/combinations           *           *        262        796  —  —  —
Meperidine/combinations           *           *           *           *  —  —  —
Methadone     1,287     1,596     1,772     3,192  —  —  —
Morphine/combinations        714     1,210           *     1,690  —  —  —
Oxycodone/combinations     5,340     4,229     7,842     9,351  — 121  —
Propoxyphene/combinations     1,888     2,129     2,811     1,754  —  —  —
Nonsteroidal anti-inflammatory agents   19,114   14,117   15,956   18,810  —  —  —
Ibuprofen   13,609   10,917   12,064   14,057  —  —  —
Naproxen     4,383     3,224     3,726     3,438  —  —  —
Salicylates/combinations     6,211     4,645     5,400     5,976  —  —  —
Miscellaneous pain medications/combinations   22,864   22,692   27,371   32,968   44   45  —
Acetaminophen/combinations   20,701   21,017   25,312   29,861   44   42  —
Tramadol/combinations     1,742     1,515     1,719     2,816  —  —  —
Tramadol     1,528     1,079     1,372     2,669  — 147  —
Acetaminophen-tramadol           *           *           *           *  —  —  —
Pain medication combinations NTA           *           *           *           *  —  —  —
Anorexiants           *           *        654           *  —  —  —
Anticonvulsants   10,957     9,389   12,580   11,803  —  —  —
Antiemetic/antivertigo agents           *           *           *        343  —  —  —
Anti-Parkinson agents          80        543           *        755  —  —  —
General anesthetics           *           *           *           *  —  —  —
Muscle relaxants     5,921     5,785     7,072     9,772   65   69   38
Carisoprodol     1,864     2,038     3,811     4,301 131 111  —
Cyclobenzaprine     2,966     2,784     2,096     3,839  —  —  —
Miscellaneous CNS agents           *           *           *           *  —  —  —
RESPIRATORY AGENTS     8,361     7,662     8,415   10,175  —  —  —
Antihistamines     2,059     1,650     1,627     3,813  — 131 134
Bronchodilators           *           *           *           *  —  —  —
Decongestants           *           *     1,347        805  —  —  —
Expectorants           *        474     1,068        649  —  —  —
Upper respiratory combinations     4,818     4,207     3,982     4,067  —  —  —
Respiratory agents NTA           *     1,244        660     1,114  —  —  —
CARDIOVASCULAR AGENTS     7,667     5,814     7,965     7,873  —  —  —
Antiadrenergic agents, centrally acting        995        912     1,929        790  —  —  -59
Beta-adrenergic blocking agents     2,105     1,916     1,999     2,501  —  —  —
Calcium channel blocking agents        879        193     1,040        601  —  —  —
Diuretics           *        539           *     1,360  —  —  —
Cardiovascular agents NTA     3,661     3,024     3,298     4,344  —  —  —


SEEKING DETOX SERVICES

ED visits involving seeking detox services, 2007

The category of visits referred to as "seeking detox" includes various situations such as nonemergency requests for admission for detox, visits to obtain medical clearance before entry to a detox program,15 and acute emergencies in which an individual is in distress (i.e., displaying active withdrawal symptoms) and seeking detox. As detox may be sought through other avenues (e.g., direct admission to a hospital, services provided through private clinics, entry into programs outside the community), the overall demand for detox services is most likely higher than suggested by DAWN estimates.

DAWN estimates 139,908 (CI: 110,901 to 168,915) drug-related ED visits for patients seeking detoxification or substance abuse treatment services during 2007 (Table 25). Two thirds (66%) of the seeking detox ED visits involved multiple drugs, and more than one third (41%) of all seeking detox ED visits involved alcohol. However, the role of alcohol may be underrepresented here because, for patients aged 21 and older, DAWN captures alcohol use only when it is combined with the use of other drugs.

Table 25
ED visits involving seeking detox services, by selected drugs, 2007
Drug category and selected drugs (1) ED visits (2,3) Percent of ED visits (3) RSE (%) 95% CI: Lower bound 95% CI: Upper bound
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(3) ED visits often involve multiple drugs. Such visits will appear multiple times in this table (e.g., a visit involving both cocaine and marijuana will appear twice in this table). The sum of visits or rates by drug will be greater than the total, and the sum of percentages by drug will be greater than 100.
NOTE: CI = confidence interval. CNS = central nervous system. NOS = not otherwise specified. NTA = not tabulated above. RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, seeking detox 139,908 100.0 10.6 110,901 168,915
Visits involving a single drug   48,159   34.4 12.0   36,788   59,531
Visits involving multiple drugs   91,748   65.6 11.4   71,270 112,226
Alcohol   57,157   40.9 12.7   42,943   71,371
Alcohol in combination   56,574   40.4 12.8   42,360   70,788
Alcohol alone           *       * 51.6           *           *
Non-alcohol illicits 106,660   76.2 11.4   82,857 130,463
Cocaine   65,124   46.5 12.7   48,889   81,359
Heroin   42,242   30.2 11.5   32,701   51,782
Marijuana   25,970   18.6 17.2   17,212   34,728
Stimulants     7,161     5.1 34.1     2,376   11,947
Amphetamines        979     0.7 27.6        450     1,508
Methamphetamine     6,287     4.5 38.4     1,549   11,025
MDMA (Ecstasy)        654     0.5 29.6        275     1,033
GHB           *       *     *           *           *
Flunitrazepam (Rohypnol)           *       *     *           *           *
Ketamine           *       *     *           *           *
LSD           *       * 50.5           *           *
PCP           *       * 70.2           *           *
Miscellaneous hallucinogens           *       *     *           *           *
Inhalants           *       *     *           *           *
Combinations NTA        216     0.2 43.5          32        399
PSYCHOTHERAPEUTIC AGENTS   21,669   15.5 13.0   16,136   27,202
Antidepressants     1,314     0.9 33.8        444     2,184
SSRI antidepressants        360     0.3 34.8        114        606
Antipsychotics        536     0.4 27.8        244        827
Atypical antipsychotics        416     0.3 29.7        174        657
Anxiolytics, sedatives, and hypnotics   20,365   14.6 13.1   15,145   25,585
Barbiturates        722     0.5 31.4        278     1,166
Benzodiazepines   19,301   13.8 13.1   14,338   24,265
Alprazolam     9,138     6.5 17.7     5,967   12,308
Clonazepam     2,635     1.9 17.1     1,754     3,516
Diazepam     3,172     2.3 22.7     1,758     4,586
Lorazepam     1,980     1.4 29.9        821     3,140
Benzodiazepines NOS     4,736     3.4 20.6     2,827     6,646
Misc. anxiolytics, sedatives, and hypnotics     1,136     0.8 32.1        422     1,850
Zolpidem        574     0.4 46.6          50     1,099
CNS stimulants     1,049     0.7 47.5          73     2,025
CNS AGENTS   43,219   30.9 12.6   32,536   53,901
Pain medications   42,776   30.6 12.6   32,179   53,373
Opiates/opioids   41,241   29.5 12.7   30,967   51,514
Opiates/opioids, unspecified     4,746     3.4 23.3     2,581     6,912
Narcotic pain medications   37,040   26.5 12.9   27,707   46,372
Fentanyl/combinations     1,359     1.0 28.3        605     2,114
Hydrocodone/combinations   10,425     7.5 17.3     6,886   13,965
Methadone     6,886     4.9 21.5     3,979     9,793
Morphine/combinations     3,341     2.4 42.2        577     6,105
Oxycodone/combinations   18,880   13.5 15.9   12,989   24,771
Nonsteroidal anti-inflammatory agents           *       *     *           *           *
Miscellaneous pain medications/combinations     2,128     1.5 23.8     1,136     3,121
Tramadol/combinations        858     0.6 33.6        293     1,422
Tramadol        858     0.6 33.6        293     1,422
Pain medications NOS        590     0.4 42.5          98     1,082
Anticonvulsants        263     0.2 43.3          40        486
Muscle relaxants     1,701     1.2 25.2        862     2,540
Skeletal muscle relaxants     1,521     1.1 25.3        766     2,275
Carisoprodol     1,108     0.8 27.4        512     1,704
CARDIOVASCULAR AGENTS        632     0.5 38.2        159     1,106
Antiadrenergic agents, centrally acting        251     0.2 49.6            7        495
Clonidine        251     0.2 49.6            7        495

Among the illicit drugs, cocaine was observed in 47 percent of visits, heroin in 30 percent, marijuana in 19 percent, and stimulants in 5 percent. Among central nervous system agents, narcotic pain medications were observed in 26 percent of visits, including oxycodone in 14 percent, hydrocodone in 8 percent, and methadone in 5 percent. Benzodiazepines were observed in 14 percent of visits.

When population size and the margin of error are taken into account, the rate of seeking detox visits for males (61 per 100,000 population) was higher than that for females (32 per 100,000 population) (Table 26, Figure 7). Rates of seeking detox visits peaked at 80 or more persons per 100,000 for those aged 18 to 44, with lower levels found for younger and older patients.

Table 26
ED visits involving seeking detox services, by patient demographics, 2007
Patient demographics ED visits (1) Percent of ED visits Rates of ED visits per 100,000 population (2)
(1) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(2) All rates are ED visits per 100,000 population. Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States. Population estimates are drawn from the 2007 U.S. Census Bureau Postcensal Resident Population National Population Dataset as of July 1, 2007.
NOTE: RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed. A dash (—) indicates a blank cell. Rates are not provided for race and ethnicity subgroups because of data limitations.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, seeking detox 139,908 100.0 45.9
Gender         —     —   —
Male   91,065   65.1 60.6
Female   48,748   34.8 31.6
Unknown           *       *     *
Age         —     —   —
0–5 years           *       *     *
6–11 years           *       *     *
12–17 years     3,369     2.4 13.4
18–20 years   10,431     7.5 80.6
21–24 years   15,180   10.8 89.7
25–29 years   21,065   15.1 98.1
30–34 years   16,370   11.7 83.0
35–44 years   39,086   27.9 91.7
45–54 years   26,622   19.0 60.0
55–64 years     6,468     4.6 19.2
65 years and older     1,293     0.9   3.3
Unknown           *       *     *
Race/ethnicity         —     —   —
White   82,105   58.7   —
Black   31,811   22.7   —
Hispanic   11,693     8.4   —
Other or two or more race/ethnicities        868     0.6   —
Unknown   13,431     9.6   —

Figure 7
Rates of ED visits per 100,000 population involving seeking detox services, by age and gender, 2007

Figure 7   D

SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).

In terms of race/ethnicity, the majority (59%) of seeking detox visits involved patients who were white. Unfortunately, DAWN is unable to produce population-based rates for race/ethnicity categories. Race/ethnicity information in ED records is often missing or is very limited. By necessity, DAWN uses a simplified set of race/ethnicity categories that is incompatible with the categories used by the U.S. Census Bureau to report population by race/ethnicity. Therefore, the population denominators that would enable DAWN to produce rates are not available.

Patients' dispositions after ED visits involving seeking detox are displayed in Table 27. Nearly a third (30%) of patients were released with a referral to a detox or treatment program, about a quarter (22%) were admitted to the detox unit in the hospital, and a smaller portion (9%) were transferred to other facilities. Some type of follow-up care was received by nearly 7 out of 10 (69%) patients who entered the ED seeking detox services.

Table 27
ED visits involving seeking detox services, by patient disposition, 2007
Patient disposition ED visits (1) Percent of ED visits Rates of ED visits per 100,000 population (2)
(1) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(2) All rates are ED visits per 100,000 population. Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States. Population estimates are drawn from the 2007 U.S. Census Bureau Postcensal Resident Population National Population Dataset as of July 1, 2007.
NOTE: RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, seeking detox 139,908 100.0 45.9
Treated and released   76,880   55.0 25.2
Discharged home   34,785   24.9 11.4
Released to police/jail           *       *     *
Referred to detox/treatment   41,832   29.9 13.7
Admitted to this hospital   42,566   30.4 14.0
ICU/critical care     1,860     1.3   0.6
Surgery           *       *     *
Chemical dependency/detox   30,542   21.8 10.0
Psychiatric unit     5,240     3.7   1.7
Other inpatient unit     4,899     3.5   1.6
Other follow-up   20,461   14.6   6.7
Transferred   12,351     8.8   4.1
Left against medical advice     2,578     1.8   0.8
Died           *       *     *
Other           *       *     *
Not documented        739     0.5   0.2

Trends in ED visits involving seeking detox services, 2004–2007

This section presents the trends in the estimates of ED visits involving seeking detox services for the period 2004 through 2007 (Table 28). Differences between years are presented in terms of the percentage increase or decrease in visits in 2007 compared with the estimates for the previous 3 years. Only statistically significant changes are discussed and displayed in the table.

Table 28
Trends in ED visits involving seeking detox services, by selected drugs, 2004–2007
Drug category and selected drugs (1) ED visits, 2004 (2,3) ED visits, 2005 (2,3) ED visits, 2006 (2,3) ED visits, 2007 (2,3) Percent change 2004, 2007 (4) Percent change 2005, 2007 (4) Percent change 2006, 2007 (4)
(1) The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
(2) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(3) ED visits often involve multiple drugs. Such visits will appear multiple times in this table (e.g., a visit involving both cocaine and marijuana will appear twice in this table). The sum of visits or rates by drug will be greater than the total, and the sum of percentages by drug will be greater than 100.
(4) This column denotes statistically significant (p < 0.05) increases or decreases between estimates for the periods shown.
NOTE: CNS = central nervous system. NTA = not tabulated above. RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed. A dash (—) indicates a blank cell.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total ED visits, seeking detox services 141,867 126,226 118,355 139,908 18
Alcohol   53,662   47,494   47,102   57,157
Alcohol in combination   51,831   47,154   46,769   56,574
Alcohol alone           *           *           *           *
Non-alcohol illicits 110,792 101,244   92,385 106,660
Cocaine   62,989   56,061   57,738   65,124
Heroin   47,035   40,895   34,462   42,242
Marijuana   25,965   22,486   22,104   25,970
Stimulants   11,760   15,402     8,128     7,161
Amphetamines           *           *     2,034        979
Methamphetamine           *           *     6,211     6,287
MDMA (Ecstasy)        882        511        483        654
GHB           *           *           *           *
Flunitrazepam (Rohypnol)           *           *           *           *
Ketamine           *           *           *           *
LSD           *           *           *           *
PCP        827        729        989           *
Miscellaneous hallucinogens           *           *           *           *
Inhalants           *           *           *           *
Combinations NTA           *        191           *        216
PSYCHOTHERAPEUTIC AGENTS   16,929   17,833   17,903   21,669
Antidepressants     1,024     1,195     1,141     1,314
SSRI antidepressants        716           *        365        360
Antipsychotics        459        259        457        536
Atypical antipsychotics        429        226        329        416
Anxiolytics, sedatives, and hypnotics   15,748   16,533   16,799   20,365
Barbiturates        852        684        530        722
Benzodiazepines   14,717   15,734   15,801   19,301
Alprazolam     6,061     6,253     7,063     9,138
Clonazepam     1,510     1,805     2,119     2,635 74
Diazepam     2,975     2,058     1,431     3,172 122  
Lorazepam     1,012        987     1,479     1,980
Misc. anxiolytics, sedatives, and hypnotics        818        751        783     1,136
CNS stimulants           *        829        589     1,049
CNS AGENTS   35,451   30,820   32,385   43,219 40 33
Pain medications   34,730   30,114   31,690   42,776 42 35
Opiates/opioids   33,296   29,330   30,786   41,241 41 34
Opiates/opioids, unspecified     4,507     4,246     4,467     4,746
Narcotic pain medications   29,894   25,550   26,880   37,040 45 38
Codeine/combinations        650        347        426           *
Fentanyl/combinations        704     1,265     1,054     1,359
Hydrocodone/combinations     8,114     8,929     8,092   10,425
Hydromorphone/combinations        962        617           *           *
Methadone     8,109     4,172     5,294     6,886 65
Morphine/combinations     1,638     2,399     3,002     3,341
Oxycodone/combinations   15,917   14,028   14,721   18,880
Propoxyphene/combinations     1,059           *        830           *
Miscellaneous pain medications/combinations     1,307     1,044     1,069     2,128 99
Acetaminophen/combinations     1,115           *        486           *
Tramadol/combinations           *        486        375        858
Anticonvulsants        455          97           *        263
Muscle relaxants     1,356     1,204     1,214     1,701

The number of patients seeking detox services through the ED was relatively stable from 2004 through 2007; the 18 percent increase from 2006 to 2007 simply brought the number of patients seeking detox back in line with its 2004 level. Although no change was observed in the involvement of illicit drugs (e.g., cocaine, heroin), significant changes were observed in pharmaceuticals. Visits involving opiate/opioid painkillers jumped 41 percent from 2005 to 2007 and were involved in more than 40,000 patient visits seeking detox in 2007. Narcotic painkillers in general, and hydrocodone and oxycodone in particular, were a large part of that increase. Although they are implicated in far fewer visits than opiates and opioids, two types of benzodiazepines have increased significantly: clonazepam rose 74 percent from 2004 to 2007 (2,635 visits in 2007), and diazepam was up 122 percent from 2006 to 2007 (3,172 visits in 2007).


APPENDIX A

MULTUM LEXICON
END-USER LICENSE AGREEMENT

1. Introduction

  1. This License Agreement (the License) applies to the Multum Lexicon database (the Database). This License does not apply to any other products or services of Cerner Multum, Inc. (Multum). A work based on the Database means either the Database or any derivative work under copyright law; i.e., a work containing the Database or a substantial portion of it, either verbatim or with modifications. A translation of the Database is included without limitation in the term modification. Each end-user/licensee is addressed herein as you.

  2. Your use of the Database acknowledges acceptance of these restrictions, disclaimers, and limitations. You expressly acknowledge and agree that Multum is not responsible for the results of your decisions resulting from the use of the Database, including, but not limited to, your choosing to seek or not to seek professional medical care, or from choosing or not choosing specific treatment based on the Database.

  3. Every effort has been made to ensure that the information provided in the Database is accurate, up-to-date, and complete, but no guarantee is made to that effect. In addition, the drug information contained herein may be time sensitive.

  4. Multum does not assume any responsibility for any aspect of healthcare administered or not administered with the aid of information the Database provides.

2. Terms and Conditions for Copying, Distribution and Modification

  1. You may copy and distribute verbatim copies of the Database as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty; keep intact all the notices that refer to this License and to the absence of any warranty; and give any other recipients of the Database a copy of this License (the readme.txt file) along with the Database and anything else that is part of the package, which should be identified.

  2. You may modify your copy or copies of the Database or any portion of it to form a derivative work, and copy and distribute such modifications or work under the terms of Section 2.A. above, provided that you also meet all of these conditions:

    1. You must cause the modified files to carry prominent notices stating that they are derived from the Multum Lexicon database from Cerner Multum, Inc. and that you changed the files and the date of any change(s).

    2. If you incorporate modified files into a computer program, you must cause it, when started running for interactive use in the most ordinary way, to print or display an announcement including an appropriate copyright notice, a notice that you have modified the Multum Lexicon database from Cerner Multum, Inc., and a notice that there is no warranty (or that you provide the warranty) and telling the user how to view a copy of this License.

  3. It is not the intent of this section to claim rights or contest your rights to work written entirely by you; rather, the intent is to exercise the right to control the distribution of derivative or collective works based on the Database.

  4. You may copy and distribute the Database (or a work based on it, under Section 2.B.) in an encoded form under the terms of Sections 2.A. and 2.B. above provided that you also do one of the following:

    1. Accompany it with the complete corresponding machine-readable plain text, which must be distributed under the terms of Sections 2.A and 2.B. above on a medium customarily used for software interchange; or,

    2. Accompany it with a written offer to give any third party, for no charge, a complete machine-readable copy of the Database (and the entirety of your derivative work based on it, under Section 2.B.), to be distributed under the terms of Sections 2.A. and 2.B. above on a medium customarily used for software interchange.

  5. You may not copy, modify, sublicense, or distribute the Database except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Database will automatically terminate your rights under this License. However, parties who have received copies, or rights, from you under this License will not have their licenses terminated so long as such parties remain in full compliance.

  6. You are not required to accept this License. However, nothing else grants you permission to copy, modify or distribute the Database or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by copying, modifying or distributing the Database (or any work based on the Database), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Database or works based on it.

  7. Each time you redistribute the Database (or any work based on the Database), the recipient automatically receives a license from Multum to copy, distribute or modify the Database subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License.

3. Disclaimer of Warranties; Limitation of Damages

  1. BECAUSE THE DATABASE IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE PROGRAM OR DATA, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING. MULTUM AND/OR OTHER PARTIES PROVIDE THE DATABASE AS IS WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED, STATUTORY OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE DATABASE IS WITH YOU. SHOULD THE DATABASE PROVE DEFECTIVE, INCOMPLETE, OR INACCURATE, YOU ASSUME THE RESPONSIBILITY AND COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.

  2. IN NO EVENT (UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING) WILL MULTUM, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR REDISTRIBUTE THE DATABASE AS PERMITTED ABOVE, BE LIABLE FOR ANY SPECIAL, INCIDENTAL, CONSEQUENTIAL, OR INDIRECT DAMAGES, INCLUDING DAMAGES FOR LOSS OF PROFITS, LOSS OF BUSINESS, OR DOWN TIME, EVEN IF MULTUM OR ANY OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.

  3. IN ADDITION, WITHOUT LIMITING THE FOREGOING, THE DATABASE HAS BEEN DESIGNED FOR USE IN THE UNITED STATES ONLY AND COVERS THE DRUG PRODUCTS USED IN PRACTICE IN THE UNITED STATES. MULTUM PROVIDES NO CLINICAL INFORMATION OR CHECKS FOR DRUGS NOT AVAILABLE FOR SALE IN THE UNITED STATES AND CLINICAL PRACTICE PATTERNS OUTSIDE THE UNITED STATES MAY DIFFER SUBSTANTIALLY FROM INFORMATION SUPPLIED BY THE DATABASE. MULTUM DOES NOT WARRANT THAT USES OUTSIDE THE UNITED STATES ARE APPROPRIATE.

  4. You acknowledge that updates to the Database are at the sole discretion of Multum. Multum makes no representations or warranties whatsoever, express or implied, with respect to the compatibility of the Database, or future releases thereof, with any computer hardware or software, nor does Multum represent or warrant the continuity of the features or the facilities provided by or through the Database as between various releases thereof.

  5. Any warranties expressly provided herein do not apply if: (i) the end-user alters, mishandles or improperly uses, stores or installs all, or any part, of the Database, (ii) the end-user uses, stores or installs the Database on a computer system which fails to meet the specifications provided by Multum, or (iii) the breach of warranty arises out of or in connection with acts or omissions of persons other than Multum.

4. Assumption of Risk, Disclaimer of Liability, Indemnity

  1. THE END-USER ASSUMES ALL RISK FOR SELECTION AND USE OF THE DATABASE AND CONTENT PROVIDED THEREON. MULTUM SHALL NOT BE RESPONSIBLE FOR ANY ERRORS, MISSTATEMENTS, INACCURACIES OR OMISSIONS REGARDING CONTENT DELIVERED THROUGH THE DATABASE OR ANY DELAYS IN OR INTERRUPTIONS OF SUCH DELIVERY.

  2. THE END-USER ACKNOWLEDGES THAT MULTUM: (A) HAS NO CONTROL OF OR RESPONSIBILITY FOR THE ENDUSER'S USE OF THE DATABASE OR CONTENT PROVIDED THEREON, (B) HAS NO KNOWLEDGE OF THE SPECIFIC OR UNIQUE CIRCUMSTANCES UNDER WHICH THE DATABASE OR CONTENT PROVIDED THEREON MAY BE USED BY THE END-USER, (C) UNDERTAKES NO OBLIGATION TO SUPPLEMENT OR UPDATE CONTENT OF THE DATABASE, AND (D) HAS NO LIABILITY TO ANY PERSON FOR ANY DATA OR INFORMATION INPUT ON THE DATABASE BY PERSONS OTHER THAN MULTUM.

  3. MULTUM SHALL NOT BE LIABLE TO ANY PERSON (INCLUDING BUT NOT LIMITED TO THE END-USER AND PERSONS TREATED BY OR ON BEHALF OF THE END-USER) FOR, AND THE END-USER AGREES TO INDEMNIFY AND HOLD MULTUM HARMLESS FROM ANY CLAIMS, LAWSUITS, PROCEEDINGS, COSTS, ATTORNEYS' FEES, DAMAGES OR OTHER LOSSES (COLLECTIVELY, LOSSES) ARISING OUT OF OR RELATING TO (A) THE END-USER'S USE OF THE DATABASE OR CONTENT PROVIDED THEREON OR ANY EQUIPMENT FURNISHED IN CONNECTION THEREWITH AND (B) ANY DATA OR INFORMATION INPUT ON THE DATABASE BY END-USER, IN ALL CASES INCLUDING BUT NOT LIMITED TO LOSSES FOR TORT, PERSONAL INJURY, MEDICAL MALPRACTICE OR PRODUCT LIABILITY.

5. Miscellaneous

  1. You warrant that you have authority within the organization you identified during registration for the Database to enter into license agreements with other organizations including Multum.

  2. You agree that Multum may identify you and/or your organization by name as a licensee, licensed user, or licensing organization of the Database or a client of Multum in Multum's external market communications. You also agree that Multum may issue, if it desires, a press release stating that you and/or your organization have licensed the Database.

  3. If conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License. If you cannot distribute so as to satisfy simultaneously your obligations under this License and any other obligations, then as a consequence you may not distribute the Database at all.

  4. If any portion of this License is held invalid or unenforceable under any particular circumstance, the balance of this License is intended to apply and the License as a whole is intended to apply in other circumstances.

  5. If the distribution and/or use of the Database is or becomes restricted in certain countries either by patents or by copyrighted interfaces, Multum may add an explicit geographical distribution limitation excluding those countries, so that distribution is permitted only in or among countries not thus excluded. In such case, this License incorporates the limitation as if written in the body of this License.

Multum Lexicon © 2008

Cerner Multum, Inc.
Colorado Center Tower One
2000 South Colorado Boulevard
Suite 11000
Denver, Colorado 80222

Document revised November 30, 2000
License downloaded 9/1/2007 from: http://www.multum.com/license.htm


APPENDIX B

GLOSSARY OF TERMS

This glossary defines terms used in data collection activities, analyses, and publications associated with the emergency department (ED) component of the Drug Abuse Warning Network (DAWN).

Accidental ingestion: This category of drug-related ED visits includes those involving the accidental use of a drug, for example, childhood drug poisonings and individuals who take the wrong medication by mistake.

Adverse reaction: This category of drug-related ED visits represents the consequences of using a prescription or over-the-counter pharmaceutical for therapeutic purposes and includes visits related to adverse drug reactions, side effects, drug-drug interactions, and drug-alcohol interactions. Adverse reactions that involve a pharmaceutical with an illicit drug are excluded from this category.

Alcohol use: Alcohol is reportable for all patients when present in combination with one or more other reportable substances. For patients under the age of 21, alcohol is also reportable if it is used alone with no other substance or reportable drug. (See Drug misuse and abuse and Underage drinking.)

Case description: A description of how the drug(s) was related to the patient's ED visit. The case description, in conjunction with other documentation in the ED medical record, is used to determine if the ED visit is reportable to DAWN. It is copied verbatim from the patient's chart when possible.

Case type: See Type of case.

Case type other: See Drug misuse and abuse.

Confidence interval (CI): An interval estimate, that is, a range of values around a point estimate that takes sampling error into account. The accepted standard of confidence is 95 percent. Technically, a 95 percent CI means that, if repeated samples were drawn from the same population of hospitals using the same sampling and data collection procedures, the true population value would fall within the confidence interval 95 percent of the time. Practically, a 95 percent CI summarizes both the estimate and its margin of error in a straightforward way with a reasonable degree of confidence.

Diagnosis: The condition(s) for which the patient was treated as determined by the clinician after study.

Disposition: The location or facility to which an ED patient was referred, transferred, or released.

Treated and released includes three categories:

Admitted to this hospital includes five categories of inpatient units:

Other follow-up includes five categories:

Drug: A substance that was recorded in a DAWN case report. Substances reportable to DAWN include alcohol, illicit drugs, prescription and over-the-counter pharmaceuticals, dietary supplements, and nonpharmaceutical inhalants. DAWN publications use the term drug to refer to any of these substances. Multiple substances (drugs) can be reported for each DAWN case. Therefore, the total number of drugs exceeds the total number of DAWN cases reported.

Drug category: A generic grouping of related pharmaceuticals or other substances reported to DAWN, based on the classification system developed by Multum Information Services, a subsidiary of the Cerner Corporation, and modified for use with DAWN. (More information on the Multum system is available at http://www.multum.com/.) In general, the Multum categories reflect the therapeutic uses for prescription and over-the-counter pharmaceuticals.

Drug misuse and abuse: A group of ED visits defined broadly to include all visits associated with illicit drugs, alcohol use in combination with other drugs, alcohol use alone among those younger than 21 years, and nonmedical use of pharmaceuticals. (See also Alcohol use, Illicit drug use, Nonmedical use of pharmaceuticals, and Underage drinking.)

Drug-related ED visit: This category includes any ED visit related to recent drug use. To be a DAWN case, a drug needs only to be implicated in the visit; the drug does not have to have caused the visit. One patient may make repeated visits to an ED or to several EDs, thus producing a number of visits. The number of unique patients involved in the reported drug-related ED visits cannot be estimated, because no direct patient identifiers are collected by DAWN.

There are some circumstances in which ED visits are not reviewed for DAWN. These include persons who left before being seen by a physician, visits for suture removal, and direct admission to the hospital through the ED for women in labor.

Estimate: A statistical estimate is the value of a parameter (such as the number of drug-related ED visits) for the universe that is derived by applying sampling weights to data from a sample. Estimates of drug-related ED visits are calculated by applying weights and adjustments to the data provided by the sampled hospitals participating in DAWN. The sampling weights reflect the probability of selection; separate adjustment factors account for nonresponse, data quality, and the known total of ED visits delivered by the universe of eligible hospitals as identified by the American Hospital Association (AHA) for the relevant time period.

Hospital emergency department (ED): To be eligible for DAWN, hospitals must be non-Federal, short-stay, general medical and surgical facilities that operate one or more EDs 24 hours a day, 7 days a week. They must be located in the United States. Specialty hospitals, hospital units of institutions, long-term care facilities, pediatric hospitals, hospitals operating part-time EDs, and hospitals operated by the Veterans Health Administration and the Indian Health Service are excluded. The universe of EDs is identified from the AHA's Annual Survey Database. Participation in DAWN is limited to hospitals that meet the eligibility criteria for DAWN. (See also Universe.)

Illicit drug use: This category of drug-related ED visits includes all visits related to the use of illicit or illegal drugs. Additional clarification is provided for the following drug categories:

Malicious poisoning: See Nonmedical use of pharmaceuticals.

Metropolitan area: An area comprising a relatively large core city or cities and the adjacent geographic areas. Conceptually, these areas are integrated economic and social units with a large population center. Unless otherwise noted, metropolitan area analyses prepared by DAWN use the boundaries established by the Office of Management and Budget (OMB), as updated in 2003.

Nonmedical use of pharmaceuticals: Nonmedical use of pharmaceuticals includes taking more than the prescribed dose of a prescription pharmaceutical or more than the recommended dose of an over-the-counter pharmaceutical or supplement; taking a pharmaceutical prescribed for another individual; deliberate poisoning with a pharmaceutical by another person; and documented misuse or abuse of a prescription drug, an over-the-counter pharmaceutical, or a dietary supplement. Nonmedical use of pharmaceuticals may involve pharmaceuticals alone or pharmaceuticals in combination with illicit drugs or alcohol. Nonmedical use of pharmaceuticals includes prescription and over-the-counter pharmaceuticals in ED visits that are of the following types of cases:

(See also Drug misuse and abuse and Type of case.)

Not otherwise specified (NOS): The catch-all category for substances that are not specifically named but are qualified as a DAWN case. Terms are classified into an NOS category only when assignment to a more specific category is not possible on the basis of information in the source documentation (ED patient charts).

Not tabulated above (NTA): This designation is used when drugs or drug categories are not explicitly listed in a table. Low-incidence drugs (or drug categories) falling under a broader drug classification may be summarized into a single row under that classification and labeled as NTA.

Overmedication: See Nonmedical use of pharmaceuticals.

Oversampling: Without oversampling, one would expect a sample to resemble the population from which it was drawn. Oversampling implies the deliberate selection of a much higher proportion of certain types of sampling units than would normally be obtained in a simple, random sample. The deliberate selection of certain types of sample units is done to improve the precision of estimates of the properties of these types of sampling units. This is a form of stratified sampling. (See also Sampling, Sampling frame, and Sampling unit.)

p-value: A measure of the probability (p) that the difference between two estimates could have occurred by chance, if the estimates being compared were really the same. The larger the p-value, the more likely the difference could have occurred by chance. For example, if the difference between two DAWN estimates has a p-value of 0.01, it means that there is a 1 percent probability that the difference observed could be due to chance alone.

Population: See Universe.

Precision: The extent to which an estimate agrees with its mean value in repeated sampling. The precision of an estimate is measured inversely by its standard error (SE) or relative standard error (RSE). In DAWN publications, estimates with RSEs greater than 50 percent are regarded as too imprecise to be published. ED table cells where such estimates would have appeared contain the asterisk symbol (*). (See also Relative standard error.)

Race/ethnicity: Race/ethnicity data in DAWN are collected retrospectively from the medical record. Patients are never interviewed to obtain DAWN data. DAWN follows OMB protocol for collection of race/ethnicity when self-identification of race/ethnicity by the individual is not possible. This approach involves a single question listing six race/ethnicity groups (plus not documented) and allows for multiple responses.16 For reporting, DAWN collapses the reported race/ethnicity information into four mutually exclusive categories plus an unknown category as follows:

Race/ethnicity is missing from ED patient records about 10 percent of the time. Detail about multiple races/ethnicities may be lacking as well. Rates of ED visits per 100,000 are not calculated for race/ethnicity categories because of these data limitations.

Rate: A measure of the incidence of drug-related ED visits per 100,000 population. A rate can be calculated for the total population or for any subset defined by characteristics such as age and gender.

Relative standard error (RSE): A measure of an estimate's relative precision. The RSE of an estimate is equal to the estimate's standard error (SE) divided by the estimate itself. For example, an estimate of 2,000 cocaine visits with an SE of 200 visits has an RSE of 0.1 and is multiplied by 100 to change it to a percentage. This resulting RSE percent value is 10 percent. The larger the RSE, the less precise the estimate. Estimates with an RSE of 50 percent or greater are not published by DAWN. (See also Precision.)

Sampling: Sampling is the process of selecting a proper subset of elements from the full population so that the subset can be used to make inference to the population as a whole. A probability sample is one in which each element has a known and positive chance (probability) of selection. A simple random sample is one in which each member has the same chance of selection. In DAWN, a sample of hospitals is selected to make inference to all hospitals; DAWN uses simple random sampling within strata.

Sampling frame: A list of units from which the ED sample is drawn. All members of the sampling frame have a known probability of being selected. A sampling frame is constructed such that there is no duplication and each unit is identifiable. Ideally, the sampling frame and the universe are the same. The sampling frame for the DAWN hospital ED sample is derived from the AHA's Annual Survey Database.

Sampling unit: A member of a sample selected from a sampling frame. For the DAWN sample, the units are hospitals, and data are collected for drug-related ED visits at the responding hospitals selected for the sample.

Sampling weights: Numeric coefficients used to derive population estimates from a sample by adjusting for deviations from the original sample design due to unequal probability sampling, variable nonresponse, and other potential sources of bias.

Seeking detox: This category of drug-related ED visits captures patients seeking substance abuse treatment, drug rehabilitation, or medical clearance for admission to a drug treatment or detoxification unit. Included are nonemergency requests for admission for detox as well as acute emergencies in which an individual is in distress (i.e., displaying active withdrawal symptoms) and seeking detox.

Single-drug case: An ED visit in which only one drug was involved. DAWN collects single-drug ED visits involving alcohol alone only if the patient was less than 21 years of age.

Statistically significant: A difference between two estimates is said to be statistically significant if the value of the statistic used to test the difference is larger or smaller than would be expected by chance alone. For DAWN ED estimates, a difference is considered statistically significant if the p-value is less than 0.05. (See also p-value.)

Strata (plural), stratum (singular): Subgroups of a universe within which separate ED samples are drawn. Stratification is used to increase the precision of estimates for a given sample size, or, conversely, to reduce the sample size required to achieve the desired level of precision. The DAWN ED sample is stratified into metropolitan area cells plus an additional cell for the remainder of the United States. To ensure thorough coverage within metropolitan areas, the universe of hospitals in each is allocated into substrata identified by (1) two types of hospital ownership (public, private) and (2) up to four size categories (measured in terms of annual ED visits). This allocation creates up to eight substrata in each metropolitan area stratum. Hospitals in the stratum that covers the rest of the United States are stratified first by Census region, type of ownership, and size (also measured in terms of ED visits). A systematic sample is selected from each of the geographic strata.

Suicide attempt: This type of drug-related ED visit captures suicide attempts (e.g., attempted suicide, tried to kill self) that are documented in the medical record and in which a drug was involved. Suicidal gestures, thoughts, or ideation, including attempts to harm oneself, are not included in this category.

Type of case: A classification used to define similar DAWN cases for analysis. Each case must be assigned a type and may not be assigned more than one type. Cases are classified into one of the following eight categories: suicide attempt, seeking detox, alcohol only (age younger than 21), adverse reaction, overmedication, malicious poisoning, accidental ingestion, and other. The case is coded into the first group that meets the inclusion criteria for that group; for example, a patient 34 years of age with hives who took aspirin and no other drug would be classified into the adverse reaction group because the case did not qualify as a suicide attempt, seeking detox, or alcohol only (age younger than 21) case.

Underage drinking: This category of drug-related ED visits includes those in which alcohol was the only drug involved and the patient was younger than 21 years old.

Universe: The entire set of units for which generalizations are drawn. The universe for the DAWN ED sample is all non-Federal, short-stay, general medical and surgical hospitals in the United States that operate one or more EDs 24 hours a day, 7 days a week. Specialty hospitals, hospital units of institutions, long-term care facilities, pediatric hospitals, hospitals operating part-time EDs, and hospitals operated by the Veterans Health Administration and the Indian Health Services are excluded. The universe of EDs is identified from the AHA's Annual Survey Database.


APPENDIX C

DAWN DATA COLLECTION AND STATISTICAL METHODS

Introduction

The Drug Abuse Warning Network (DAWN) is a public health surveillance system that has monitored drug-related emergency department (ED) visits to hospitals since the early 1970s. DAWN was initially established by the Drug Enforcement Administration. Then DAWN was transferred to the U.S. Department of Health and Human Services (HHS), where the National Institute on Drug Abuse (NIDA) conducted DAWN from 1980 to 1992. Since 1992, the Office of Applied Studies (OAS) of the Substance Abuse and Mental Health Services Administration (SAMHSA), HHS, has been responsible for DAWN operations and reporting.

Since its inception, DAWN has relied on data collected from a sample of hospitals. However, over the years, the exact survey methodology has been adjusted to improve the quality, reliability, and generalizability of the information produced by DAWN. When NIDA assumed responsibility for DAWN in 1980, implementation of a sample of hospitals to produce representative estimates for the Nation and for selected metropolitan areas became a priority. This sample, refreshed with annual maintenance, continued to support DAWN estimates for the contiguous United States and 21 metropolitan areas until 2002. By that time, major population shifts and changes in the hospital industry over the preceding two decades made apparent the need for a redesign of the sample of hospitals, which was undertaken as part of a wholesale redesign of most major features of DAWN.

Currently, the DAWN survey relies on a longitudinal probability sample of hospitals located throughout the United States, including Alaska and Hawaii. To be eligible for selection into the DAWN sample, a hospital must be a non-Federal, short-stay, general surgical and medical hospital located in the United States, with at least one 24-hour ED. This current approach was first implemented in the 2004 data collection year.

DAWN uses the data from the visits classified as DAWN cases in the selected hospitals to calculate various estimates of drug-related ED visits for the Nation as a whole, as well as for specific metropolitan areas. To calculate these estimates and measure their precision requires the application of sampling and weighting methodologies to the DAWN survey.

This appendix documents the data collection methods and the sampling, weighting, and variance estimation methodologies used to develop estimates for the DAWN data collected for 2007. Additional detail on data collection methodology is available in the ED Reference Guide.17

Target population

The target population is drug-related ED visits in non-Federal, short-stay, general surgical and medical hospitals with 24-hour EDs in the United States.

Hospital sample frame

DAWN uses the American Hospital Association (AHA) Annual Survey Database as the basis for its sampling frame. The AHA maintains an updated national registry of U.S. hospitals that is estimated to have a coverage rate of 99 percent.18 A health care facility must meet several criteria to be classified as a hospital by the AHA. These criteria include the provision of patient services, diagnostic or therapeutic, for general or specific medical conditions; licensed medical staff; and accreditation by organizations such as the Joint Commission on Accreditation of Health Care Organizations. A hospital is considered to be eligible for inclusion in the DAWN sampling frame if it is a non-Federal, short-stay, general surgical and medical hospital in the United States that operates at least one 24-hour ED. Many DAWN hospitals operate multiple EDs.

Determination of DAWN eligibility

A hospital is considered ineligible if any one of the key criteria that define eligibility (non-Federal, short-stay, general surgical and medical hospital, located in the United States, 24-hour ED) is not met. Only those hospitals that meet all the criteria are considered eligible. If information for any criterion is missing (and a hospital is otherwise eligible considering the nonmissing criteria information), other variables in the AHA Annual Survey Database are used to determine eligibility. If the hospital's eligibility remains unknown, additional data sources are consulted to determine eligibility.

DAWN data collection

DAWN ED data are collected through a retrospective review of ED medical records for patients treated in the ED. Patients or families are never interviewed. The review of source records is performed by a trained DAWN Reporter in each member facility. Depending on the needs of the facility, the DAWN Reporter may be an employee of the hospital or an employee of the DAWN operations contractor.

For each facility that participates in DAWN, the designated DAWN Reporter reviews all medical records to find ED visits related to drug use. The DAWN Reporter submits an electronic case report to the DAWN system for each ED visit that meets the specific case selection criteria. DAWN Reporters also track, on a copy of the ED registration log, their progress in reviewing the universe of ED visits.

Because of the volume in some EDs, a sample of medical records is obtained rather than reviewing all charts. This subsampling introduces another component of variance that is accounted for in the weighting and estimation process.

Data items collected by DAWN

The case report form showing all the collected DAWN data items is provided in Figure C1.

Figure C1
DAWN ED case form

Figure C1   D

DAWN features that enhance data quality and reliability

Several methods are used to improve the quality and reliability of DAWN data, including the following:

ED visits eligible for DAWN

A DAWN case is any ED visit related to recent drug use. DAWN includes ED visits associated with substance abuse and misuse, both intentional and accidental. DAWN also includes ED visits related to the use of drugs for legitimate therapeutic purposes. To be a DAWN case, the relation between the ED visit and the drug need not be causal; the drug needs only to be implicated in the visit.

The case criteria are intended to be broad and inclusive and to have few exceptions. Broad criteria take into account the fact that documentation in medical records varies in clarity and comprehensiveness across hospitals and among clinicians within hospitals. Broad criteria minimize the potential for judgments that could cause data to vary systematically and unexpectedly across reporters and hospitals. In addition, broad criteria are designed to capture a very diverse set of drug-related visits that can be aggregated or disaggregated to serve a variety of analytical purposes and the interests of multiple audiences. In DAWN, only recent drug use is included, the reason a patient used a drug is irrelevant, and the criteria are broad enough to encompass all types of drug-related events, including but not limited to explicit drug abuse.

There are a few clearly delineated exceptions to the DAWN eligibility criteria. An ED visit is not a DAWN visit if

Types of cases in DAWN

By design, DAWN's broad case criteria yield a diverse set of visits. To bring order to this heterogeneous mix of ED visits, each visit is assigned to one of eight types, which may be analyzed separately or in purposeful combinations. The eight types of visits are

DAWN Reporters assign each DAWN case to one, and only one, of the eight types of cases, on the basis of a series of questions and decision rules. The questions and rules are organized into the DAWN ED Decision Tree (Figure C2). Starting at the top, each case is assigned to the first type of case that applies, even if the case might also meet the rules for a subsequent category. The eight types of case were ordered with this process in mind.

Figure C2
Type of case decision tree

Figure C2   D

The final category in the decision tree, Other, is reserved for DAWN visits that do not meet any of the rules for classification into one of the first seven types. By design, most cases of drug abuse are classified as Other. This approach, which never directly identifies drug abuse, comes from the recognition that medical records frequently lack explicit documentation of substance abuse. This lack of documentation may occur for several reasons. First, the distinctions among use, misuse, and abuse are often subjective. Second, if there is a low index of suspicion for drug abuse in some types of patients, ED physicians may be unlikely to label those types of patients as drug abusers. Third, in many States, insurers may legally deny payment for ED visits related to substance abuse. Thus, financial incentives may be a powerful factor to influence documentation practices.

Drugs included in DAWN

DAWN includes all types of drugs:19

To be reportable, a nonpharmaceutical substance must be consumed by inhalation, sniffing, or snorting, and it must have a psychoactive effect when inhaled. An ED visit involving inhalation of a nonpharmaceutical, psychoactive substance and no other drug qualifies as a DAWN case. Carbon monoxide is excluded from the inhalants. Since 2004, cases involving accidental exposures (e.g., exposure to paint fumes while one is painting a closet) have been excluded as well.

Hospital participation

For 2007, 207 hospitals submitted data on 300,983 drug-related ED visits that were used for estimation (Tables C1 and C2). The overall weighted response rate was 29.6 percent. For the 12 oversampled metropolitan areas and divisions, individual response rates ranged from 30.7 percent in the Houston-Baytown-Sugar Land, TX, Metropolitan Statistical Area to 76.3 percent in the Detroit-Warren-Livonia, MI, Metropolitan Statistical Area.

Table C1
Data collection year 2007
Geographic area Total eligible hospitals (1) Eligible hospitals in sample (1) Responding hospitals in sample Response rate for sampled hospitals (%) Design weight response rate (%) Visits weighted response rate (%)
(1) Non-Federal, short-stay hospitals with 24-hour EDs in the United States, as identified by the American Hospital Association Annual Survey, are eligible for DAWN.
(2) The total number of eligible hospitals includes the sampled and participating hospitals from metropolitan areas shown in this table plus hospitals in the remainder of the United States. Components shown here do not sum to the total.
(3) Metropolitan Statistical Areas (MSAs) and Metropolitan Divisions follow the standard definitions issued by the Office of Management and Budget in June 2003 (available at http://www.whitehouse.gov/omb/bulletins/b03-04.html), with one exception: for New York, geographic coverage is limited to the subarea comprising the five Boroughs of New York City.
NOTE: A dash (—) indicates a blank cell.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total United States (2) 4,575 542 207 38.2 25.5 29.6
Metropolitan Statistical Areas (3)     —   —   —    —    —    —
Boston-Cambridge-Quincy, MA-NH, MSA      41   29   18 62.1 62.2 64.0
Chicago-Naperville-Joliet, IL-IN-WI, MSA      89   73   30 41.1 41.8 39.0
Denver-Aurora, CO, MSA      16   15     9 60.0 60.0 65.0
Detroit-Warren-Livonia, MI, MSA      37   25   16 64.0 67.5 76.3
Houston-Baytown-Sugar Land, TX, MSA      50   42   13 31.0 32.6 30.7
Minneapolis-St. Paul-Bloomington, MN-WI, MSA      27   27   10 37.0 37.0 39.8
Phoenix-Mesa-Scottsdale, AZ, MSA      28   26   13 50.0 50.0 47.4
San Diego-Carlsbad-San Marcos, CA, MSA      16   16     7 43.8 43.8 43.1
Seattle-Tacoma-Bellevue, WA, MSA      23   23     8 34.8 34.8 44.7
Metropolitan Divisions and Subareas (3)     —   —   —    —    —    —
Miami-Fort Lauderdale-Miami Beach, FL, MSA—Dade County Division      22   16     9 56.3 51.7 59.6
New York-Newark-Edison, NY-NJ-PA, MSA—Five Boroughs Division      50   39   21 53.8 48.2 58.0
San Francisco-Oakland-Fremont, CA, MSA—San Francisco Division      18   18     8 44.4 44.4 57.5
Table C2
Drug-related ED visits and drugs, by type of case, 2007
Type of case (1) Unweighted sample data Weighted estimates RSE (%) 95% CI:
Lower bound
95% CI:
Upper bound
(1) Refer to Figure C2 for a description of types of cases and to Table C3 for a description of the relationship between types of cases and categories of drug misuse and abuse.
(2) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
(3) These are estimates of drugs involved in ED visits. Because a single ED visit may involve multiple drugs, the number of drugs is greater than the number of visits.
NOTE: CI = confidence interval. RSE = relative standard error. A dash (—) indicates a blank cell.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Drug-related ED visits (2)            —                —    —            —            —
Suicide attempt      13,913        197,053   8.4    164,564    229,542
Seeking detox      22,253        139,908 10.6    110,901    168,915
Alcohol only (age < 21)      11,278        135,900   8.4    113,470    158,331
Adverse reaction    109,258     1,908,928   9.5 1,552,364 2,265,491
Overmedication      23,554        331,134   6.8    287,226    375,043
Malicious poisoning           863          12,563 16.0        8,616      16,510
Accidental ingestion        4,872          91,632   9.2      75,087    108,177
Other    114,992     1,181,110 13.7    864,972 1,497,248
Total drug-related ED visits    300,983     3,998,228   6.7 3,473,733 4,522,724
Total drug misuse or abuse ED visits    177,989     1,883,272   8.7 1,561,490 2,205,054
Total ED visits (all reasons) 9,486,510 116,486,292   0.0            —            —
Drugs (3)            —                —    —            —            —
Suicide attempt      30,532        428,570   9.4    349,861    507,279
Seeking detox      47,622        293,133 10.9    230,639    355,627
Alcohol only (age < 21)      11,278        135,900   8.4    113,470    158,331
Adverse reaction    143,599     2,604,412 11.6 2,014,658 3,194,166
Overmedication      41,754        597,284   6.7    518,286    676,283
Malicious poisoning        1,544          22,180 15.9      15,259      29,100
Accidental ingestion        6,548        128,042 12.0      98,002    158,083
Other    196,558     2,038,197 11.7 1,572,451 2,503,943
Drugs in all drug-related ED visits    479,435     6,247,718   7.3 5,355,391 7,140,045
Drugs in all misuse or abuse ED visits    314,911     3,335,487   7.8 2,828,492 3,842,482

Charts reviewed for drug-related ED visits

DAWN cases are found through a retrospective review of medical records in participating hospitals. Across all participating hospitals in 2007, 10,413,928 charts were reviewed to find the drug-related ED visits that met the DAWN case criteria. On the basis of the review of charts, 375,030 drug-related visits were found and submitted to the DAWN database, a case rate of 3.6 percent.20 On average, a DAWN member hospital submitted 1,183 DAWN cases. However, the number of submitted cases varied widely across hospitals, from 3 cases to 6,532 cases (median 953) in a single hospital during 2007.

DAWN data in this publication

For analysis, five categories of ED visits related to drug misuse and abuse were defined. These categories were designed to parallel the approach of the National Survey on Drug Use and Health. These categories are

Two additional categories were isolated for analysis: visits involving drug-related suicide attempts and visits for the purpose of seeking detox services. Such visits are considered to be misuse or abuse only if they involve illicit drug(s) or alcohol.

These categories are defined by drug and type of case as shown in Table C3. Because multiple drugs may be involved in a single visit, these categories are not mutually exclusive. A drug-related ED visit involves, on average, about 1.6 drugs.

Table C3
DAWN analytic groups
Analytic category Drugs included Types of cases included
(1) Nonmedical use of pharmaceuticals explicitly excludes ED visits for adverse reactions and accidental ingestions.
(2) Suicide attempts and seeking detox visits are not considered to be drug misuse or abuse unless they involve an illicit drug or alcohol.
All misuse or abuse
  • All
This analytic group is the union of the following four analytic groups: use of illicit drugs, use of alcohol in combination, underage drinking, and nonmedical use of pharmaceuticals. See the definition provided for each of these groups for detail on the exact drugs and types of cases included in this overall category.
Use of illicit drugs
  • Cocaine
  • Heroin
  • Marijuana
  • Stimulants (amphetamines and methamphetamine)
  • MDMA
  • GHB
  • Flunitrazepam (Rohypnol)
  • Ketamine
  • LSD
  • PCP
  • Other hallucinogens
  • Nonpharmaceutical inhalants
  • Combinations of illicit drugs
All types of cases
Use of alcohol in combination Alcohol in combination with one or more other drugs All types of cases
Underage drinking Alcohol only, and no other drugs, in patients younger than 21 Cases with alcohol as the sole drug area are all categorized in one of the following three types of cases:
  • suicide attempts,
  • seeking detox, and
  • alcohol only (age < 21).
The patient must be younger than 21 years of age.
Nonmedical use of pharmaceuticals
  • Prescription drugs
  • Over-the-counter medications and pharmaceuticals
  • Dietary supplements
Combination of three types of cases (1):
  • overmedication (cases of nonmedical use, overuse, or misuse lacking explicit documentation of drug abuse),
  • malicious poisoning (cases in which the patient was administered a drug by another for a malicious purpose), and
  • type of case other (cases that could not be assigned to another type of case; includes documented drug abuse).
Drug-related suicide attempts (2) All drugs Must be categorized in the type of case Suicide Attempt (cases in which the records indicate the patient attempted suicide)
Visits for the purpose of seeking detox services (2) All drugs Must be categorized in the type of case Seeking Detox (cases in which the records indicate the patient was seeking detox)

Sampling and estimation

DAWN sample design

The redesign of the DAWN system that was introduced in 2003 altered most of the major features of DAWN data collection and included a new sample of hospitals that constituted DAWN. The new sampling plan, fully implemented for the first time for the 2004 estimates, formed a nationally representative panel of hospitals to be followed longitudinally for the indefinite future. The new design is a probability-based, stratified, one-stage sample. A complete and accurate list of all hospitals in the United States was drawn; from that list, all hospitals meeting the criteria for the target sample frame were identified. Samples were drawn to provide the capability to make estimates for the Nation as well as selected Metropolitan Statistical Areas (MSAs) and Metropolitan Divisions (Table C4).21 Each year the sample frame is updated to account for new hospitals.

Table C4
Oversample areas in DAWN sample design
(1) Denotes a legacy area. Two separate legacy areas (New York and Newark) are contained in the New York-Newark-Edison, NY-NJ-PA, Metropolitan Statistical Area.
Atlanta-Sandy Springs-Marietta, GA (1)
Austin-Round Rock, TX
Baltimore-Towson, MD (1)
Birmingham-Hoover, AL
Boston-Cambridge-Quincy, MA-NH (1)
Bridgeport-Stamford-Norwalk, CT
Buffalo-Cheektowaga-Tonawanda, NY (1)
Chicago-Naperville-Joliet, IL-IN-WI (1)
Cincinnati-Middletown, OH-KY-IN
Cleveland-Elyria-Mentor, OH
Columbus, OH
Dallas-Fort Worth-Arlington, TX (1)
Denver-Aurora, CO (1)
Detroit-Warren-Livonia, MI (1)
Hartford-West Hartford-East Hartford, CT
Honolulu, HI
Houston-Baytown-Sugar Land, TX
Indianapolis, IN
Kansas City, MO-KS
Los Angeles-Long Beach-Santa Ana, CA (1)
Los Angeles-Long Beach-Santa Ana, CA—Los Angeles division (contains Los Angeles-Long Beach-Glendale, CA, Metropolitan Division)
Los Angeles-Long Beach-Santa Ana, CA—Orange County division (contains Santa Ana-Anaheim-Irvine, CA, Metropolitan Division)
Las Vegas-Paradise, NV
Louisville, KY-IN
Memphis, TN-MS-AR
Miami-Fort Lauderdale-Miami Beach, FL (1)
Miami-Fort Lauderdale-Miami Beach, FL—Fort Lauderdale division (contains Fort Lauderdale-Pompano Beach-Deerfield Beach, FL, and West Palm Beach-Boca Raton-Boynton Beach, FL, Metropolitan Divisions)
Miami-Fort Lauderdale-Miami Beach, FL—Miami-Dade County division (contains Miami-Miami Beach-Kendall, FL, Metropolitan Division)
Minneapolis-St. Paul-Bloomington, MN-WI (1)
Nashville-Davidson–Murfreesboro, TN
New Haven-Milford, CT
New Orleans-Metairie-Kenner, LA (1)
New York-Newark-Edison, NY-NJ-PA (1)
New York-Newark-Edison, NY-NJ-PA—New Jersey division (contains Middlesex, Monmouth, Ocean, Somerset, Essex, Hunterdon, Morris, Sussex, Union, Bergen, Hudson, Passaic Counties, NJ, and Pike County, PA)
New York-Newark-Edison, NY-NJ-PA—New York Suburban division (contains Nassau, Putnam, Rockland, Suffolk, Westchester Counties, NY)
New York-Newark-Edison, NY-NJ-PA—New York City, 5 Boroughs division (contains Bronx, Kings, New York, Queens, Richmond Counties, NY)
Omaha-Council Bluffs, NE-IA
Philadelphia-Camden-Wilmington, PA-NJ-DE-MD (1)
Phoenix-Mesa-Scottsdale, AZ (1)
Pittsburgh, PA
Portland-Vancouver-Beaverton, OR-WA
Providence-New Bedford-Fall River, RI-MA
Riverside-San Bernardino-Ontario, CA
Rochester, NY
Sacramento-Arden-Arcade-Roseville, CA
Salt Lake City, UT
San Antonio, TX
San Diego-Carlsbad-San Marcos, CA (1)
San Francisco-Oakland-Fremont, CA (1)
San Francisco-Oakland-Fremont, CA—Oakland division (contains Oakland-Fremont-Hayward, CA, Metropolitan Division)
San Francisco-Oakland-Fremont, CA—San Francisco division (contains San Francisco-San Mateo-Redwood City, CA, Metropolitan Division)
Seattle-Tacoma-Bellevue, WA (1)
St. Louis, MO-IL (1)
Tampa-St. Petersburg-Clearwater, FL
Tucson, AZ
Washington-Arlington-Alexandria, DC-VA-MD-WV (1)
Wichita, KS

The stratified design called for drawing oversamples of hospitals in 48 MSAs; in 4 of those 48 MSAs, additional oversamples were drawn for a total of nine divisions. In effect, 53 nonoverlapping geographic areas (44 whole MSAs and 9 divisions) were in the sampling frame. (See Table C4 for list of MSAs and divisions where oversamples were drawn.) These areas are collectively referred to as oversample areas, or OS areas.

Metropolitan Statistical Areas and Divisions

To accommodate a planned expansion of the metropolitan areas covered by DAWN, a maximum set of metropolitan areas, based on the definitions issued by the Office of Management and Budget (OMB) in June 2003, was selected. Which metropolitan areas to include was a topic of the DAWN redesign.22 Retention of the existing 21 metropolitan areas was important because significant demand existed for estimates for those areas, and addition of the five most populous metropolitan areas in each of the nine Census divisions was deemed important to improve DAWN's geographic and population coverage. This decision yielded a total of 48 metropolitan areas. For many of the 48 metropolitan areas, the June 2003 definitions resulted in larger metropolitan areas. In some cases, these larger areas represented a merger of previously separate metropolitan areas. However, users of DAWN statistics continued to be strongly interested in the areas covered by the original 21 metropolitan areas. To address the needs of these users, four of the merged areas were subdivided.23 For each of these areas, there was a sample for the metropolitan area, as well as a sample for each division. This enables DAWN to produce estimates for the metropolitan areas and for the divisions. The final metropolitan-area sample included a total of 53 geographic units: 48 metropolitan areas, 2 divisions each for 3 of these metropolitan areas, and 3 divisions for one of these metropolitan areas.

Sampled hospitals in each of the OS areas were stratified by hospital size (up to four categories on the basis of volume of ED visits) and ownership type (public and private). (Size categories were determined independently for each OS area.) The stratification plan included an additional geographic construct for the remainder of the United States outside the OS areas. Hospitals in the remainder area were stratified into 24 remainder strata on the basis of four regions (Northeast, South, Midwest, West), hospital size (three size categories by volume of ED visits), and ownership type (public and private).

To begin, a cross classification was created by categories of ownership type and geographic unit. Within each combination of geographic area and ownership type, the number of hospitals determined the number of unique size categories. If there were three or fewer hospitals, only one size category was defined. If there were four, five, six, or seven hospitals, two size categories were defined. If there were eight or more hospitals, four size categories were defined. In the remainder sample, within each combination of Census region and ownership, there were three size categories. This produced 24 unique strata from which to draw the hospitals for the remainder sample.24

The DAWN national estimates are the sum of the estimates for OS areas and the remainder area. Using a formula, the national estimate is depicted as

Equation C1   D

where ai is the estimate for OS area i, 53 is the number of OS areas, and b is the remainder area estimate.

It was never expected that DAWN would be able to expand data collection into all 53 OS areas. Instead, the expectation was that DAWN would build up gradually to the number of OS areas its budget could support. The DAWN sample design was conceived to provide the flexibility to change gradually over time in terms of the number of OS areas where data were collected, while providing the statistical infrastructure to enable the production of reliable and representative estimates for the Nation and selected OS areas, regardless of their number.

To accomplish this objective, the DAWN design incorporates an approach whereby a subset of the hospitals within the OS areas was identified a priori as having a dual purpose in estimation. Referred to as dual-purpose hospitals, these designated hospitals can contribute to an estimate for the OS area in which they are located or they can contribute to the estimate for the remainder area. Dual-purpose hospitals carry two probabilities of selection (POS) and two stratum identifiers. One POS/stratum is associated with membership in an OS-area sample, and the other is associated with membership in the remainder-area sample.25

Figure C3 depicts the initial sample as it was drawn to provide

Figure C3
Original DAWN sample design

Figure C3   D

For estimation for each data year, the first step is to determine which role each sampled hospital will play in that year's estimates. To make that determination, the response rates and nonresponse patterns for each OS area are reviewed to determine data quality. Those OS areas with acceptable data quality are allowed to stand on their own as the basis for separate estimates; they are referred to as stand-alone OS areas. All hospitals in stand-alone OS areas, including those originally designated as being in the dual-purpose subsample, are considered to be oversample hospitals in the OS areas, and they contribute to the OS-area estimate using their OS-area POS/stratum.

If it is determined on the basis of response rates and bias analyses that an OS area cannot stand alone, the design provides that the OS area is eliminated as a separate area but becomes part of the remainder area. In this instance

Figure C4 depicts the assignment of dual-purpose hospitals to either an OS area or the remainder area and the exclusion of OS hospitals outside of stand-alone OS areas that are not designated as dual purpose.

Figure C4
DAWN design in practice

Figure C4   D

After it is determined which OS areas will stand alone, the DAWN national estimates as reported in this publication are the sum of the estimates for stand-alone OS areas plus the remainder area. Using a formula, the national estimate is depicted as

Equation C2   D

where ai is the estimate for stand-alone OS area i, N is the number of stand-alone OS areas, and b is the remainder area estimate inclusive of dual-purpose hospitals in OS areas that do not stand alone.

It is important to note that the definitions of the remainder area and the remainder sample of hospitals are designed to be fluid; hospital membership in the remainder sample changes from year to year depending on the response rates and data quality within the OS areas.

Sample maintenance

Because DAWN is a longitudinal survey that will be used to analyze trends in drug-related ED visits over time, annual updates to the sample are performed to ensure that the sample remains representative of the target population. The initial sample was selected in 2003 from a sampling frame created from the 2001 AHA Annual Survey Database. In every subsequent year, the sampling frame is updated to reflect new, closed, merged, and demerged hospitals, on the basis of updates to the AHA files. These updates include newly eligible hospitals, which are new hospitals or previously ineligible hospitals that are now eligible. Each year, the newly eligible hospitals are provided the opportunity to be selected into the sample, on the basis of the sampling fraction of the stratum in which the newly eligible hospital is located.

Reduction of bias

Survey error is the extent to which findings from the survey sample differ from those of the population of interest. The statistical methodologies described above are designed to minimize error. Additional sources of error, often referred to as bias, also contribute to overall error. Measuring bias is difficult because it requires accurate knowledge about corresponding population values. The DAWN survey methodology includes proven techniques, practices, and protocols that reduce the potential for introducing bias. For example, clearly defined criteria are used to construct the initial hospital sampling frame. Coverage bias is minimized because the sampling frame has virtually 100 percent coverage of the target population. To minimize measurement bias, the individuals who collect data for DAWN are provided with specialized and intensive training, automated methods for data entry are used, and the data are subject to quality reviews at several points in the data collection process. Additional details on the survey data collection methodologies used to enhance DAWN data quality and reduce bias are provided in an earlier DAWN publication.26

Sample size and sample allocation

DAWN defines precision in terms of the relative standard error (RSE) of an estimate. The RSE is the standard error of the estimate divided by the actual point estimate. DAWN is designed to have RSEs less than or equal to 10 percent for metropolitan-area estimates and RSEs less than or equal to 15 percent for national estimates pertaining to total drug-related visits, cocaine visits, heroin visits, and marijuana visits. As discussed below, these desired precision levels are important drivers for setting sample size targets.

Sample sizes for each geographic area were determined by the area's targeted precision level in combination with the theory of optimal allocation for stratified samples. According to this approach, the variance of the sample estimates will be minimized when the sample size, nh, in each sampling stratum is made proportional to the quantity:

Equation C3   D

where Wh is the proportion of sampling units, Sh is the population standard deviation for the parameter being measured, and Ch represents the square root of the cost of sampling in stratum h.

Using these optimum allocation conditions, the minimum required sample sizes necessary to achieve the targeted levels of precision in each DAWN area were calculated using the following general considerations:

In addition to these considerations, sampling rates (i.e., the number of sampled hospitals divided by the number of eligible hospitals) were also subject to the following constraints:

Response rate calculations

In 2007, the initial DAWN sample included 1,287 hospitals divided among 53 OS areas (48 MSAs and 9 divisions) and one remainder area. Response rates and nonresponse bias analyses were assessed to determine which of these 53 OS areas could stand alone (Figure C3). Once this determination was made, hospitals that were neither dual purpose nor located in a stand-alone OS area were treated as if they were not sampled. For 2007, this treatment has the effect of reducing the sample from 1,287 hospitals to 542 hospitals, which is the number used for purposes of computing the unweighted response rates (Table C1).

Of the 53 original OS areas, a total of 12 areas (9 metropolitan areas and 3 submetropolitan areas) were determined to be able to stand alone in 2007.

Sampling weights

The DAWN hospitals are selected using stratified simple random sampling with oversampling in selected metropolitan areas. The stratum sample sizes were determined through an optimum allocation process. Sampling weights are first calculated as the inverse of the probability of selection and then adjusted for variable nonresponse and by a procedure known as poststratification, or benchmark adjustment.

Within-hospital weighting adjustment

Within-hospital nonresponse occurs when a hospital provides incomplete data. To minimize the impact of within-hospital nonresponse, the DAWN weighting plan includes nonresponse adjustment factors that were developed and applied for each month of data collection within each facility. The within-hospital nonresponse adjustment factor is calculated as the total number of ED visits within a month within a facility divided by the total number of reviewed charts for that same facility-month.

The within-hospital weights are applied to the case data by month and by facility. That is, the visit counts for a given facility-month are first summed for each drug and then multiplied by the corresponding within-hospital adjustment factor for that facility-month. The weighted totals are then summed over all facilities and months to give a total weighted visit count for each drug for each hospital.

Weighting adjustment for hospital nonresponse

Hospital-level nonresponse occurs when hospitals fail to provide any data. To minimize the impact of hospital nonresponse, the DAWN weighting plan includes nonresponse adjustment factors that were developed and applied within each weighting class. Weighting classes were formed on the basis of the aforementioned sampling stratification schemes. Within each weighting class, the nonresponse adjustment factor is calculated as the sum of the sampled hospital weights divided by the sum of the weights of the responding hospitals. The hospital nonresponse adjustment factors are checked to make sure the adjustments are within reasonable bounds. If a nonresponse adjustment factor is out of bounds (either too small or too large), adjacent weighting classes are collapsed and new nonresponse adjustment factors are calculated.

When the hospital-level nonresponse adjustment factors were considered final, a nonresponse-adjusted sampling weight was then calculated as the product of the nonresponse adjustment factor and the sampling weight. For each weighting class, a verification check was conducted to ensure that the sum of the nonresponse-adjusted sampling weights was equal to the sum of the sampled hospital weights.

Weighting adjustment for population benchmarks (poststratification)

The DAWN weighting plan also includes a poststratification adjustment factor that reconciles the weighted number of total visits for responding hospitals with the number of total visits from the most recent AHA Annual Survey Database. DAWN used a ratio adjustment within strata to implement this adjustment.

Poststratification strata were formed on the basis of the aforementioned sampling stratification schemes. Within each stratum, the adjustment factor was calculated as the ratio of the AHA count of total visits to the weighted sum of total visits for responding hospitals. The factors were verified to ensure they were within reasonable bounds. If they were out of bounds (either too small or too large), adjacent poststratification strata were collapsed and new poststratification adjustment factors were calculated.

When the poststratification adjustment factors were considered final, a poststratified weight was then calculated. The final weight was calculated as the product of the poststratification adjustment factor and the nonresponse-adjusted sampling weight. For each poststratification stratum, a validity check was conducted to ensure that the sum of the poststratified weighted total visits was equal to the corresponding AHA count of total visits from each stratum.

Total drug-related ED visits

Estimates for the entire universe of DAWN-eligible hospitals in the United States are produced by applying poststratified weights to the data received from the sampled hospitals. Thus, for 2007, 300,983 submitted cases were extrapolated to an estimate of 3,998,228 drug-related ED visits. Considering the margin of error, this estimate may range from 3,473,733 to 4,522,724 drug-related ED visits, out of approximately 116 million total ED visits estimated for the United States (Table C2).

Calculation of estimates

All estimates produced for this publication were calculated using data that had been weighted according to the plan described above. Estimates for any variable of interest were determined by first summing the case totals within facility-month, applying the within-hospital weight, summing to the hospital level, applying the final hospital weight, and summing over all hospitals.

Variance estimation

Each hospital in the DAWN sample was selected through a random process, which theoretically could have been repeated many times, resulting in many hypothetical samples. Sampling variance, or the margin of error, refers to the extent to which these samples vary. Two measures of this variability are the standard error (SE) and the relative standard error (RSE), which is defined as the SE of the estimate divided by the estimate itself. The precision of an estimate is inversely related to the sampling variance, as measured by the RSE. The greater the RSE value, the lower the precision.

For example, if 10,000 estimated visits involve a given drug, and this estimate has an SE of 500 visits, then the RSE value is 5 percent:

RSE = SE/Estimate
RSE = 500/10,000
RSE = 0.05, or 5% when multiplied by 100 for percentage RSE.

In this publication, confidence intervals (CIs) are included in many of the tables and are often cited in the text along with the estimates. The 95 percent CI is calculated as

CI = Estimate ± (1.96 x RSE x Estimate),

where 1.96 comes from the table of normal distribution z-values and means that 95 percent of the normal distribution lies within 1.96 standard deviations of the mean.

Applying the formula to the example above, the 95 percent CI would be

10,000 ± (1.96 x 0.05 x 10,000) = 10,000 ± 980.0
Lower limit: 10,000 - 980 = 9,020
Upper limit: 10,000 + 980 = 10,980
95% CI: 9,020 to 10,980.

If repeated samples were drawn from the same population of hospitals, using the same sampling and data collection procedures, the true population values would fall within that interval 95 percent of the time.

Both between- and within-hospital variance components were accounted for. Within-hospital variance was estimated using a replication strategy by which two random replicates were created within each hospital and the variance between the two replicates represented the within-hospital contribution. Typically, this component was considerably smaller than the between-hospital variance, which was calculated as the variance between weighted hospital totals within each stratum.

Variance estimates reported in this publication were determined using Taylor Series Linearization. Variance estimates were calculated using SUDAAN® software.

Standardized rates

Standardized measures are needed to make valid comparisons of estimates across age and gender categories. For age in particular, the size of the underlying population differs considerably across age groups; for example, the number of individuals aged 18 to 20 in the United States is much lower than the number of individuals aged 35 to 44. All other factors being the same, a higher estimate of ED visits would be expected to occur naturally for the group that is larger in the population. In this example, assume that the RSE is 0.25 (25%).

To take the size of the underlying population into account, rates of ED visits per 100,000 population were calculated using population data from the U.S. Census Bureau.27

For each age and gender category, the estimate for a category was divided by the population for that category, which was then divided by 100,000. For example, consider an estimate of 1,000 visits for an age group of 1,000,000 persons, and an estimate of 1,000 visits for an age group of 500,000 persons. The rates would be calculated as

1,000 / (1,000,000/100,000) = 1,000/10
= 100 visits per 100,000 population

1,000 / (500,000/100,000) = 1,000/5
= 200 visits per 100,000 population.

Applying the formula to the example above, the 95 percent CI for the higher rate would be

200 ± (1.96 x.25 x 200) = 200 ± 98
Lower limit: 200 - 98 = 102
Upper limit: 200 + 98 = 298
95% CI for the rate: 102 to 298.

The RSE value is the same as the one used when calculating confidence interval surrounding the point estimate of ED visits.

Population estimates used to generate rates for 2007 are provided in Table C5.

Table C5
U.S. population by age and gender, 2007
Gender and age Total United States (1) Males Females
(1) Population estimates for 2007 are, as of 7/29/2008, from the U.S. Census Bureau Postcensal Resident Population National Population Dataset, National estimates by demographic characteristics—single year of age, sex, race, and Hispanic Origin, Monthly Population Estimates. Link: http://www.census.gov/popest/datasets.html. File: NC-EST2007-ALLDATA-R-File20.csv.
Total 304,482,526 150,164,126 154,318,400
0–5 years   25,073,004   12,828,002   12,245,002
6–11 years   23,861,515   12,198,787   11,662,728
12–17 years   25,069,516   12,838,986   12,230,530
18–20 years   12,946,887     6,646,831     6,300,056
21–24 years   16,914,311     8,722,806     8,191,505
25–29 years   21,463,403   11,014,453   10,448,950
30–34 years   19,712,434   10,019,562     9,692,872
35–44 years   42,614,900   21,373,849   21,241,051
45–54 years   44,387,175   21,859,897   22,527,278
55–65 years   33,669,357   16,240,765   17,428,592
65 years and older   38,770,024   16,420,188   22,349,836

Standardized rates were not calculated for race and ethnicity subgroups, because the race/ethnicity categories available to DAWN are much less detailed and contain considerably more missing data than the race and ethnicity categories in the census data. Appendix D describes the race and ethnicity data reported for DAWN.

Determination of significant differences between years

Comparisons in the estimates of ED visits between years are presented in the form of percentage differences, calculated as the 2007 estimate minus the 2004 estimate divided by the 2004 estimate. For shorter-term comparisons, these percentages are calculated as the 2007 estimate minus the 2005 estimate divided by the 2005 estimate or the 2007 estimate minus the 2006 estimate divided by the 2006 estimate. The result is presented as a percentage, which is shown only if the difference between the two years is statistically significant. Tests for the significance of differences between two years consider the variance of each year's estimate and the covariance between the two. Thus, hospitals that appear in both samples and provide data in both years contribute to the covariance and thus decrease the overall sampling variance beyond the combined contribution of the two samples. The variance estimation process used to establish significance takes into account this overlap between the two annual samples.

Publication criteria

DAWN is based on a survey and collects detailed data using more than 17,000 drug codes. As a result, some estimates will be too imprecise, too small, or based on too little data to be reliable. In these situations, the estimate is replaced by an asterisk (*) in the published table. Estimates are suppressed according to the following rules:

It is mathematically possible that an estimate could have no sampling error and an RSE of zero. This occurs when the number of ED visits being estimated is small, all the hospitals contributing to that estimate were selected with certainty, and the absence of any sampled hospital is due to nonresponse. In most cases, an estimate with an RSE of zero is suppressed on the basis of the small number of cases. In the unlikely event that an estimate is published with an RSE of zero, it is most appropriate to interpret the RSE as signifying that the necessary data were not available to approximate the sampling error.


APPENDIX D

RACE AND ETHNICITY IN DAWN

In October 1997, the Office of Management and Budget (OMB) issued a revised standard protocol for race and ethnicity categories used in Federal data collection systems.28 The new protocol permitted separate reporting of race and Hispanic ethnicity, and it incorporated the ability to capture more than one race for an individual, a few modifications in nomenclature (e.g., black was changed to black or African American), division of certain categories (Asian or Pacific Islander was split into two categories, Asian and Native Hawaiian or Other Pacific Islander), and elimination of the other category. The OMB protocol also permitted a combined format, whereby race and Hispanic ethnicity would be recorded in a single data item, which could still record multiple entries for race and/or Hispanic ethnicity. The single data item for race and ethnicity is shown in the Drug Abuse Warning Network (DAWN) emergency department (ED) case form that has been used since 2003 (Appendix C, Figure C1).

Because DAWN retrospectively collects data from medical records, missing information about race/ethnicity cannot be obtained at a later time (patients are never interviewed). Race/ethnicity is missing in about 10 percent of DAWN cases. Although OMB protocol allows for a combined format, there is still limited information for the DAWN race/ethnicity categories of Asian, American Indian or Alaska Native, Native Hawaiian or Other Pacific Islander, two race/ethnicities, and three race/ethnicities. Because of these limitations, the number of ED visits per 100,000 population (rates) are not presented in this report by race/ethnicity.

For reference, estimates of drug-related ED visits by race/ethnicity are presented in Table D1. This analysis, which is based on the most detailed coding of race/ethnicity in DAWN case reports, shows that estimates for the following categories are too small to be meaningful:

Table D1
Drug-related ED visits, by detailed race/ethnicity, 2007
Race/ethnicity ED visits (1)
(1) Estimates of ED visits are based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the United States.
NOTE: RSE = relative standard error. An asterisk (*) indicates that an estimate with an RSE greater than 50% or an estimate based on fewer than 30 visits has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2007 (08/2008 update).
Total drug-related ED visits 3,998,228
One race/ethnicity 3,974,737
White 2,511,973
Black or African American    680,913
Hispanic    367,261
Asian        6,406
American Indian or Alaska Native      39,463
Native Hawaiian or Other Pacific Islander               *
Race unknown    357,488
Two race/ethnicities      23,448
White + black or African American           848
White + Hispanic               *
White + Asian             90
White + American Indian or Alaska Native               *
Black or African American + Hispanic        1,332
Black or African American + Asian               *
Black or African American + American Indian or Alaska Native               *
Hispanic + Asian               *
Hispanic + American Indian or Alaska Native               *
Asian + American Indian or Alaska Native               *
Three race/ethnicities               *
White + black or African American + Hispanic               *
White + Hispanic + Asian               *
White + Asian + Native Hawaiian or Other Pacific Islander               *

Therefore, in the tables of estimates in this and other DAWN publications, we have retained a more limited set of categories: white, black, and Hispanic. A fourth category, called Race/ethnicity not tabulated above (NTA), is used to tabulate those categories that are too small to report independently.29 All cases reported to DAWN as Hispanic or Latino ethnicity are tabulated as Hispanic race/ethnicity, regardless of race.

End Notes

1 The 95 percent confidence interval (CI) accounts for the margin of error of the estimate. It indicates with a high degree of confidence that the true population value was between 1,561,490 and 2,205,054 drug-related ED visits.
2 Although a drug was implicated in each visit, these attempts are not limited to drug overdoses.
3 Percentages add to greater than 100 percent because visits often involve multiple drugs.
4 These visits do not represent the full extent of the demand for detox services, as many programs do not require medical clearance through the ED for program entry.
5 Table C1 in Appendix C provides detail on response rates for each metropolitan area.
6 The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A.
7 These four categories of ED visits are not mutually exclusive. The sum of visits or rates by category will be greater than the total, and the sum of percentages will be greater than 100.
8 ED patients aged 21 or older for whom alcohol was the only drug associated with their ED visits are not considered DAWN cases.
9 Heroin ED visits may be underestimated. When drugs related to an ED visit are determined through toxicology tests, often the results do not distinguish heroin from the general category of "unspecified opiates." The number of drug misuse/abuse ED visits involving unspecified opiates is estimated at 57,219 (CI: 43,872 to 70,566) visits, and about half of these (30,568) were determined through toxicology testing. What portion of these toxicology results is attributable to heroin is unknown.
10 National Institute on Alcohol Abuse and Alcoholism (NIAAA). Frequently asked questions for the general public. Retrieved November 18, 2009, from http://www.niaaa.nih.gov/FAQs/General-English/default.htm#taking_medications
11 DAWN focuses on ED visits related to recent drug use and excludes medications taken on a regular basis that are not related to the ED visit. ED records frequently do not distinguish methadone used properly for the treatment of opiate addiction (and not specifically related to the ED visit) from nonmedical methadone use (related to the ED visit). This could result in overreporting the estimated number of ED visits related to methadone, but the extent of the overreporting is unknown.
12 This includes only single-ingredient formulations. Many multi-ingredient pharmaceuticals containing diphenhydramine are classified elsewhere (e.g., as respiratory agents).
13 Excluded are suicide-related behaviors documented as something other than actual attempts (e.g., suicidal ideation, suicidal gesture, or suicidal thoughts).
14 Percentages add to greater than 100 percent because visits often involve multiple drugs.
15 Some detox programs, in the hospital or the community, require medical clearance before a person can be admitted to a program. Medical clearance establishes whether a person has any special medical needs (e.g., person is diabetic and needs insulin) or is not suitable to mingle with other patients in the program (e.g., person has an infectious disease).
16 See Office of Management and Budget, Revisions to the standards for the classification of Federal data on race and ethnicity, 62 Fed. Reg. 58,782 (October 30, 1997).
17 The ED Reference Guide is available for download from the DAWN Web site, https://dawninfo.samhsa.gov/collect/. The link for the document is https://dawninfo.samhsa.gov/files/collect_2009-2011/ed_reference_guide_2009-2011.pdf.
18 AHA Annual Survey Database, Fiscal year 2001. Health Forum LLC. Copyright 2003, One North Franklin Street, Chicago, IL 60606.
19 The classification of drugs used in DAWN is derived from the Multum Lexicon, © 2008, Multum Information Services, Inc. The classification was modified to meet DAWN's unique requirements (2008). The Multum Licensing Agreement governing use of the Lexicon is provided in Appendix A and can be found on the Internet at http://www.multum.com.
20 For 2007, more hospitals participated in DAWN than were used in estimation. Therefore, the number of drug-related ED visits from all participating hospitals exceeded the number used for estimation.
21 MSAs and Metropolitan Divisions follow the standard definitions issued by OMB in June 2003 (available at http://www.whitehouse.gov/omb/bulletins/b03-04.html and http://www.census.gov/population/www/estimates/metrodef.html), with one exception. The four MSAs where samples were drawn for divisions are Los Angeles, Miami, New York, and San Francisco. The division definitions follow OMB standards except in New York, where three submetropolitan areas were defined on the basis of local input.
22 Substance Abuse and Mental Health Services Administration, Office of Applied Studies. (2002). Drug Abuse Warning Network: Development of a new design (methodology report) (DAWN Series M-4, DHHS Publication No. SMA 02-3754). Rockville, MD: Author.
23 When metropolitan areas were redefined in June 2003, on the basis of data from the 2000 decennial Census, several legacy MSAs were merged with other MSAs to form new, much larger MSAs. However, a strong constituency of DAWN data users still needed estimates for the pre-merger areas. Because of this, 4 of the 48 metropolitan areas—Los Angeles, Miami, New York, and San Francisco—were subdivided into a total of nine divisions, corresponding to the constituents' areas of interest.
24 Four Census regions times two ownership categories times three size categories equals 24 strata.
25 In addition, a portion of hospitals in the nine oversampled divisions were identified a priori to serve in their MSA-level oversample and were assigned an OS area–level POS/stratum for that third purpose. Therefore, hospitals in the four MSAs with division-level oversampling can have up to three nonzero POS/strata: (1) a POS/stratum for membership in the MSA; (2) a POS/stratum for membership in the division; and (3) a POS/stratum for membership in the remainder area.
26 Substance Abuse and Mental Health Services Administration, Office of Applied Studies. (2004). Appendix B: Technical notes: Changes to improve the quality of DAWN data. In Drug Abuse Warning Network 2003: Interim national estimates of drug-related emergency department visits (DAWN Series D-26, DHHS Publication No. SMA 04-3972). Rockville, MD: Author.
27 Population estimates for 2007 are, as of 7/29/2008, from the U.S. Census Bureau Postcensal Resident Population National Population Dataset, National estimates by demographic characteristics—single year of age, sex, race, and Hispanic Origin, Monthly Population Estimates. Link: http://www.census.gov/popest/datasets.html. File: NC-EST2007-ALLDATA-R-File20.csv.
28 See Office of Management and Budget, Revisions to the standards for the classification of Federal data on race and ethnicity, 62 Fed. Reg. 58,782 (October 30, 1997).
29 One exception is that, if two races are reported and the second is reported as unknown, the episode is coded for the known race.

Go to Top of PageGo to the Table of Contents