Click here to go to the Table of Contents for the DAWN 2004: National Estimates of Drug-Related Emergency Department Visits

Skip to Main Content

OFFICE OF APPLIED STUDIES


Drug Abuse Warning Network, 2004:
National Estimates of Drug-Related
Emergency Department Visits



U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES
Substance Abuse and Mental Health Services Administration
http://DAWNinfo.samhsa.gov/



ACKNOWLEDGMENTS

This publication was prepared by the Substance Abuse and Mental Health Services Administration (SAMHSA), Office of Applied Studies (OAS) with RTI International under Contract No. 280-03-2602. Judy K. Ball, Ph.D., M.P.A. (DAWN Project Director, SAMHSA/OAS), Scott Novak, Ph.D. (RTI), and Darryl Creel, M.S. (RTI) wrote the publication. Other significant contributors included Victoria Albright, M.A. (Project Director, RTI), Karol Krotki, Ph.D. (RTI), Eric Johnson, Ph.D. (RTI), Francine Cannarozzi, M.Ed. (RTI), Erin Mallonee, M.S. (SAMHSA/OAS), and Elizabeth Crane, Ph.D., M.P.H. (SAMHSA/OAS). The DAWN data collection was conducted by Westat under Contract No. 283-02-9025 under the direction of Josefina Moran.

PUBLIC DOMAIN NOTICE

All material appearing in this publication is in the public domain and may be reproduced or copied without permission from the Substance Abuse and Mental Health Services Administration (SAMHSA). However, this publication may not be reproduced or distributed for a fee without the specific, written authorization of the Office of Communications, SAMHSA. Citation of the source is appreciated. Suggested citation:

Substance Abuse and Mental Health Services Administration, Office of Applied Studies. Drug Abuse Warning Network, 2004: National Estimates of Drug-Related Emergency Department Visits. DAWN Series D-28, DHHS Publication No. (SMA) 06-4143, Rockville, MD, 2006.

OBTAINING ADDITIONAL COPIES OF PUBLICATION

Copies may be obtained, free of charge, from the National Clearinghouse for Alcohol and Drug Information (NCADI). The NCADI is a service of the Substance Abuse and Mental Health Services Administration (SAMHSA). Write or call NCADI at:

National Clearinghouse for Alcohol and Drug Information (NCADI)
P.O. Box 2345, Rockville, MD 20847-2345
301-468-2600  |  1-800-729-6686  |  TDD 1-800-487-4889

ELECTRONIC ACCESS TO PUBLICATION

This publication can be accessed electronically through Internet World Wide Web connections:

http://DAWNinfo.samhsa.gov/
http://www.samhsa.gov/
http://www.oas.samhsa.gov/

ORIGINATING OFFICE

Substance Abuse and Mental Health Services Administration
Office of Applied Studies
1 Choke Cherry Road, Room 7-1044, Rockville, MD 20857
April 2006



CONTENTS

Acknowledgments

Highlights
   Total drug-related ED visits
      ED visits involving drug misuse/abuse
   Illicit drugs in ED visits
   Alcohol and drug-related ED visits
      Alcohol in combination with other drugs
      Alcohol in patients under age 21
   Non-medical use of pharmaceuticals and drug-related ED visits
   Special types of drug-related ED visits
      Suicide attempts
      Seeking detox

Introduction
   Major features of DAWN
      What is a DAWN case?
      Types of cases in DAWN
      What drugs are included in DAWN?
      Other DAWN features
   Estimates in this publication
      Hospital participation in 2004
      The margin of error
      Estimates adjusted for population size

Drug-related ED visits in 2004
   Total drug-related ED visits
   Drug-related ED visits by type of case
   Drug misuse and abuse in ED visits

Illicit drugs in ED visits

Alcohol and drug-related ED visits
   Alcohol in combination with other drugs
   Alcohol only in patients under the age of 21
   Any alcohol in patients under the age of 21

Non-medical use of pharmaceuticals and drug-related ED visits
   Non-medical use of pharmaceuticals

Special types of drug-related ED visits
   Suicide attempt
   Seeking detox

List of Tables
   Table 1. DAWN ED sample and response rates: 2004
   Table 2. Drug-related ED visits, by type of case: 2004
   Table 3. Drug-use and misuse in ED visits in the U.S., by type of drug involvement: 2004
   Table 4. Illicit drugs and alcohol in drug-related ED visits: 2004
   Table 5. Illicit drugs, by type of case: 2004
   Table 6. Illicit drugs, by patient characteristics: 2004
   Table 7. Alcohol in drug-related ED visits: 2004
   Table 8. Alcohol, by type of case: 2004
   Table 9. Drugs reported most frequently with alcohol, by type of case: 2004
   Table 10. Alcohol, by patient characteristics: 2004
   Table 11. Alcohol only (age < 21), by patient and visit characteristics: 2004
   Table 12. Alcohol in drug-related ED visits in patients under age 21: 2004
   Table 13. Non-medical use of pharmaceuticals: 2004
   Table 14. Non-medical use of pharmaceuticals, by patient and visit characteristics: 2004
   Table 15. Suicide attempt: 2004
   Table 16. Suicide attempt, by patient and visit characteristics: 2004
   Table 17. Seeking detox: 2004
   Table 18. Seeking detox, by patient and visit characteristics: 2004

List of Figures
   Figure 1. Type of case decision tree
   Figure 2. DAWN ED case form
   Figure 3. Drug-related ED visits in the U.S., by type of case: 2004
   Figure 4. Illicit drugs in ED visits: 2004
   Figure 5. Illicit drugs, ED visit rates by age and gender: 2004
   Figure 6. Alcohol with other drugs, ED visit rates by age and gender: 2004
   Figure 7. Alcohol only (age < 21), ED visit rates by age and gender: 2004
   Figure 8. Non-medical use of pharmaceuticals, ED visit rates by age and gender: 2004
   Figure 9. Suicide attempt, ED visit rates by age and gender: 2004
   Figure 10. Seeking detox, ED visit rates by age and gender: 2004

List of Appendixes
   Appendix A: Multum Lexicon End-User License Agreement
   Appendix B: DAWN Methodology
   Appendix C : Glossary of Terms
   Appendix D: Population Data
   Appendix E: Race and Ethnicity in DAWN



HIGHLIGHTS

This publication presents national estimates of drug-related visits to hospital emergency departments (EDs) for 2004, based on data from the Drug Abuse Warning Network (DAWN). These estimates pertain to the entire U.S., including Alaska, Hawaii, and the District of Columbia. The Substance Abuse and Mental Health Services Administration (SAMSHA) is the agency responsible for DAWN. SAMHSA is required to collect data on drug-related emergency department visits under section 505 of the Public Health Service Act.

DAWN estimates for 2004 are the first to be based on the new, redesigned sample of hospitals, which covers the entire U.S. Estimates for 2004 also cover a full 12-month period for the first time since the redesign of DAWN was introduced.1 Therefore, the estimates in this publication establish a new baseline against which subsequent years' estimates may be compared. No comparisons with prior years should be made.

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. Estimates for 2004 are based on data submitted by 417 hospitals. In each participating hospital, ED medical records are reviewed retrospectively to find the ED visits that were related to recent drug use. All types of drugs—illegal drugs, prescription and over-the-counter pharmaceuticals, dietary supplements, and nonpharmaceutical inhalants—are included. Alcohol, when it is the only drug implicated in a visit, is included for patients younger than age 21; alcohol, when it is present in combination with another drug, is included for patients of all ages.

Total drug-related ED visits

Of an estimated 106 million ED visits in the U.S. during 2004, DAWN estimates that 1,997,993 (95% confidence interval (CI): 1,708,205 to 2,287,781) were drug-related.2

ED visits involving drug misuse/abuse

Out of a total of nearly 2 million drug-related ED visits in 2004, DAWN estimates that nearly 1.3 million ED visits were associated with drug misuse or abuse. Of those ED visits involving drug misuse or abuse:

Illicit drugs in ED visits

For 2004, DAWN estimates 940,953 (CI: 773,124 to 1,108,782) drug-related ED visits involved a major substance of abuse. DAWN estimates that:

Taking the margin of error into account, the stimulants (amphetamines and methamphetamine) may be as frequent as heroin in drug-related ED visits, but the stimulants are less frequent than cocaine or marijuana. Since some drug screens test for amphetamines only as a class, an amphetamine-positive result could indicate amphetamine or methamphetamine.

After taking population size and the margin of error into account:

Alcohol and drug-related ED visits

DAWN estimates that, for 2004, 461,809 (CI: 375,820 to 547,798) drug-related ED visits involved alcohol in combination with another drug or alcohol alone in a patient under the age of 21. Thus, nearly a quarter (23%) of all drug-related ED visits involved alcohol in one of these forms. Since DAWN does not account for ED visits involving alcohol alone in adults, the actual number of ED visits involving alcohol is higher. Alcohol is reported to DAWN when it is present in combination with other drugs, regardless of the patient's age.

Alcohol in combination with other drugs

In 2004, DAWN estimates that 363,641 (CI: 289,516 to 437,766) ED visits involved the use of alcohol in combination with another drug. Alcohol was most frequently combined with:

Alcohol in patients under age 21

Considering ED visits only for patients under the age of 21, DAWN estimates 96,809 (CI: 76,127 to 117,491) drug-related ED visits involved alcohol and no other drug.

Injuries were diagnosed in 29% of the alcohol-only visits, and accidents involving falls or motor vehicles were diagnosed in 7%. Most (85%) of these visits resulted in patients being treated and released, usually to home; another 9% were admitted to inpatient units.

Taking population size and the margin of error into account:

Alcohol use by minors also occurs in combination with other drugs. Considering alcohol only and alcohol in combination with other drugs, DAWN estimates 60,118 (CI: 44,918 to 75,318) drug-related ED visits for patients aged 12 to 17 and 82,583 (CI: 67,853 to 97,313) drug-related ED visits for patients aged 18 to 20.

Non-medical use of pharmaceuticals and drug-related ED visits

DAWN estimates 495,732 (CI: 408,285 to 583,179) ED visits in 2004 for non-medical use—i.e., misuse or abuse—of prescription or over-the-counter (OTC) pharmaceuticals. Multiple drugs were involved in more than half (57%) of these ED visits. The most frequent drugs in these visits were central nervous system (CNS) agents (53% of visits) and psychotherapeutic agents (48% of visits).

Among the CNS agents, the most frequent drugs were opiate/opioid analgesics (32% of visits involving non-medical use), including single-ingredient (e.g., oxycodone) and combination forms (e.g., hydrocodone with acetaminophen). Methadone and single-ingredient and combination forms of oxycodone and hydrocodone were the most frequent opioids, occurring in similar numbers of visits:

It is not possible to know the extent to which the source of these drugs is a legitimate prescription versus other sources nor is it possible to distinguish methadone used for treatment of opiate addiction from the methadone in pill form that is prescribed for pain.

Among the psychotherapeutic agents, the anxiolytics (anti-anxiety agents), sedatives, and hypnotics are the most frequent, occurring in more than a third (35%) of visits associated with pharmaceutical misuse/abuse. ED visits involving benzodiazepines clearly outnumber those involving any of the other types of psychotherapeutic agents. DAWN estimates that 144,385 (CI: 115,520 to 173,250) ED visits associated with pharmaceutical misuse/abuse involved benzodiazepines in 2004. This is comparable to the number for opiates/opioids.

Taking population size and the margin of error into account:

Special types of drug-related ED visits

Suicide attempts

DAWN estimates 121,585 (CI: 108,955 to 134,215) drug-related ED visits associated with suicide attempts for 2004. The majority of suicide attempt ED visits involved multiple drugs (64%).

In these ED visits for drug-related suicide attempts in 2004:

Seeking detox

DAWN estimates 177,879 (CI: 70,845 to 284,913) drug-related ED visits for patients seeking detox or substance abuse treatment services during 2004. However, these visits tend to be concentrated in hospitals with administrative policies that require medical clearance in the ED for admission to these specialized units.

More than 60% of ED visits for seeking detox involved multiple drugs. Both illicit and prescription drugs were common in these visits:

Among the seeking detox ED visits, 7 out of 10 received some type of follow-up care, either inpatient admission, referral elsewhere for detox or substance abuse treatment services, or transfer to another health care facility. However, a quarter of seeking detox cases may not have received the care they sought because they were discharged to home.

INTRODUCTION

This publication presents final estimates of drug-related emergency department (ED) visits from the Drug Abuse Warning Network (DAWN) for 2004. DAWN is a public health surveillance system that monitors drug-related emergency department (ED) visits for the nation and for selected metropolitan areas. DAWN also collects data on drug-related deaths investigated by medical examiners and coroners in selected metropolitan areas and States. 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.

Major changes to DAWN were instituted at the beginning of 2003. These changes are the result of a redesign that, among other improvements, altered most of DAWN's core features, including the design of the hospital sample and the cases eligible for DAWN. These improvements create a permanent disruption in trends. As a result, comparisons cannot be made between old DAWN (2002 and prior years) and the new DAWN.

This publication presents national estimates of drug-related ED visits for 2004, based on data collected between January 1, 2004 and December 31, 2004. These are final estimates and the first full-year estimates from the new DAWN. Therefore, no trends are presented in this publication.

The findings based on the new DAWN hospital sample are representative of the entire United States, and, as such, they are generalizable to all 50 States and the District of Columbia.

Major features of DAWN

What is a DAWN case?

One of the most important features of DAWN is its expansive definition of a case:

A DAWN case is any ED visit related to recent drug use.

DAWN includes ED visits associated with substance abuse/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. DAWN cases are found through a retrospective review of medical records.3 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 cases, which can be aggregated and disaggregated to serve a variety of analytical purposes and the interests of multiple audiences. In DAWN, only recent drug use is included;4 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 case criteria. An ED visit is not a DAWN case if:

Types of cases in DAWN

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

Each DAWN case is assigned hierarchically into one and only one case type based on a series of questions and rules. To assign case type, DAWN reporters use a decision tree, a graphical depiction of the logic of the case type assignment rules (Figure 1). Cases are classified into the first case type that applies. Even if a case might fit into more than one type, it is assigned to the first one that applies. The case types were ordered with this in mind.

 

Figure 1
Type of case decision tree

Figure 1   D

The final category, the case type called other, is reserved for DAWN cases 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 case type 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 of drugs 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 factor to influence documentation practices.

With these eight case types DAWN includes some ED visits that are unrelated to drug abuse. However, using the hierarchical decision tree is a method for isolating a set of cases involving drug abuse or misuse.

What drugs are included in DAWN?

DAWN includes all types of drugs.5 Drugs in DAWN include:

To be reportable, a non-pharmaceutical 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 non-pharmaceutical, psychoactive substance and no other drug qualifies as a DAWN case. Carbon monoxide is excluded from the inhalants. Beginning in 2004, cases involving accidental exposures (e.g., exposure to paint fumes while painting a closet) are excluded as well.

Other DAWN features

Several methods are used to improve the quality and reliability of DAWN data. These include:

The case report form showing all the DAWN data items is provided in Figure 2.

Estimates in this publication

Estimates in this publication were calculated from a probability sample of hospitals by applying sampling weights to data from the sample and accounting for the survey design. Only national estimates pertaining to the U.S. are provided.

Estimates for 2004 are, for the first time, representative of the entire 50 States and the District of Columbia. Hospitals eligible for the DAWN sample are non-Federal, short-stay, general, surgical and medical hospitals in the U.S. that operate 24-hour EDs. The American Hospital Association's (AHA's) 2001 Annual Survey is the source of the sampling frame. (For a definition of sampling frame and other technical terms used in this publication, see Appendix C, Glossary of Terms.)

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 U.S., 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.

For 2004, the national estimates are calculated as the sum of the estimates from 16 geographic areas (15 metropolitan areas, divisions, and subareas and the remainder of the U.S.). The sampling weights consist of three components:

The nonresponse adjustment to the sampling weights is designed to account for data that are missing, but not for data that are incomplete. Therefore, the data used for this publication were subjected to an intense level of scrutiny. The procedures observed for 2004 differed somewhat from those applied in 2003 due to the increased volume of data. For 2004, DAWN case eligibility and assignment of type of case were subjected to a multi-stage review involving automated "expert system" processing with selective manual verification. First, each record submitted as a DAWN case was reviewed by an expert system, which assigned a probability that the record met DAWN case criteria. Records not meeting minimum probability thresholds, as well as a subsample of those that did, were reviewed by DAWN staff for final case eligibility determinations. Second, the expert system reviewed data items submitted on each DAWN case and assigned a probability for each case type. The case type with the highest probability was compared with the case type originally reported on the record. When these agreed, the case was flagged as final. When these disagreed or when the case type probability did not meet a minimum threshold, the case was reviewed manually to resolve the differences. An additional 10% of cases were reviewed manually as a quality control check. Third, all data were checked for internal consistency, out-of-range values, missing data, and adherence to skip patterns at data entry and during subsequent cleaning processes.

 

Figure 2
DAWN ED case form

Figure 2   D

A fourth and final review focused explicitly on the issue of incomplete data, that is, DAWN cases missed due to incomplete chart review or inappropriate application of the case criteria. This review used statistical process control methods and information gained from on-site quality audits to identify and evaluate unexpected variability across months in the number of medical charts reviewed and the number of DAWN cases submitted for each hospital.

Hospital participation in 2004 (Table 1)

For 2004, 417 hospitals submitted data that were used for estimation. The weighted response rate varied from 47.4% in the San Francisco Division of the San Francisco Metropolitan Statistical Area (MSA) to 78.1% in the Buffalo, NY MSA. The weighted response rate for hospitals covering the U.S. outside of the 15 metropolitan areas, divisions, and subareas was 35.3%.

Across the 417 participating hospitals, more than 12 million charts were reviewed to find the drug-related visits that met the DAWN case criteria. Based on the review of charts, 279,564 drug-related visits were found and submitted. On average, a DAWN member hospital submitted 670 DAWN cases. However, the number of cases varied widely, from 4 cases to 7,485 (median 402) in a single hospital during 2004.

The margin of error

Since DAWN relies on a sample of hospitals, each estimate produced from the DAWN ED sample data is subject to sampling variability, the so-called "margin of error." This 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), which is defined as the standard error of the estimate expressed as a percentage of the value of the estimate. The precision of an estimate is inversely related to its sampling variability as measured by the RSE. The greater the RSE, the lower the precision.

DAWN estimates with RSE values greater than 50% and estimates less than 30 are regarded as too imprecise for publication and are not shown. In the tables, three dots ("…") are shown in the place of estimates that have an RSE greater than 50% or estimates less than 30. Ratios (percentages or rates per 100,000 population) based on suppressed estimates are likewise suppressed. Gray shading in a cell indicates that the cell is not applicable. For example, no drugs other than alcohol can be present in the "alcohol only" case type category.

In this publication, confidence intervals (CIs) are included in many of the tables and are cited in the text along with the estimates. A CI, which is expressed as a range of values, does a better job of reflecting the true nature of the statistical estimates because it takes both the estimate and its margin of error into account. A 95% 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% of the time.

For readers unfamiliar with these concepts, a more detailed discussion and examples are provided in Appendix B.

 

Table 1
DAWN ED sample and response rates: 2004
Geographic area Total eligible hospitals1 Eligible hospitals in sample Responding hospitals in sample Response rate for sample hospitals Response rate for visits (weighted)
Total U.S.2 4,438 951 417 43.8 47.6
Metropolitan Statistical Areas (MSAs)3
Atlanta-Sandy Springs-Marietta, GA MSA 41 31 15 48.4 57.8
Boston-Cambridge-Quincy, MA-NH MSA 41 30 16 53.3 59.6
Buffalo-Cheektowaga-Tonawanda, NY MSA 14 14 8 57.1 78.1
Chicago-Naperville-Joliet, IL-IN-WI MSA 91 75 34 45.3 47.5
Denver-Aurora, CO MSA 14 14 8 57.1 65.0
Detroit-Warren-Livonia, MI MSA 38 26 20 76.9 72.6
New Orleans-Metairie-Kenner, LA MSA 21 21 10 47.6 68.9
Phoenix-Mesa-Scottsdale, AZ MSA 25 25 11 44.0 52.5
St. Louis, MO-IL MSA 37 37 17 45.9 49.1
San Diego-Carlsbad-San Marcos, CA MSA 17 17 10 58.8 61.4
Seattle-Tacoma-Bellevue, WA MSA 22 22 12 54.5 54.9
Washington-Arlington-Alexandria, DC-VA-MD-WV MSA 34 30 14 46.7 53.0
Metropolitan Divisions and Subareas3
Miami-Miami Beach-Kendall, FL
Metropolitan Division of Miami-Fort Lauderdale-
Miami Beach, FL MSA
21 17 11 64.7 68.9
Bronx, Kings, New York, Queens,
Richmond Counties of New York-
Newark-Edison, NY-NJ-PA MSA
52 40 26 65.0 75.1
San Francisco-San Mateo-Redwood City, CA
Metropolitan Division of San Francisco-
Oakland-Fremont, CA MSA
18 18 9 50.0 47.4
1 Short-term, general, non-Federal hospitals with 24-hour emergency departments, based on the American Hospital Association (AHA) Annual Survey, are eligible for DAWN.
2 Total eligible hospitals in the U.S. include eligible hospitals from metropolitan areas shown and the remainder of the U.S. Therefore, components shown 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.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

Estimates adjusted for population size

Standardized measures are needed to make valid comparisons of ED visits and drugs across age and gender categories that differ in population size. 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 U.S. is much lower than the number of individuals aged 35 to 44.

To take the size of the underlying population into account, rates of ED visits per 100,000 people are generated using population data from the U.S. Bureau of the Census.6 An example is provided in Appendix B, and the population estimates used for this publication can be found in Appendix D.

Standardized rates are not calculated for race and ethnicity subgroups because the race and 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 E describes the race and ethnicity data reported to DAWN.

DRUG-RELATED ED VISITS IN 2004

Total drug-related ED visits (Table 2)

Estimates for the entire universe of DAWN-eligible hospitals in the U.S. are produced by applying sampling weights to the data received from the sampled hospitals. Thus, for 2004, 279,564 submitted cases are extrapolated to an estimate of 1,997,993 drug-related ED visits. Considering the margin of error, this estimate may range from 1,708,205 to 2,287,781 drug-related ED visits out of nearly 106 million total ED visits estimated for the U.S.

On average, a drug-related ED visit involved 1.6 drugs.

Drug-related ED visits by type of case (Figure 3)

The distribution of drug-related ED visits across the eight case types is illustrated in Figure 3. Estimates for the U.S. show the largest number of cases (35%) fell into the category other. Adverse reaction, which accounted for 30% of drug-related ED visits, is second in frequency, followed by overmedication (12%). Patients seeking detox accounted for 9% of drug-related ED visits. Suicide attempt, which was narrowly defined, accounted for 6% of drug-related visits. Visits associated with underage alcohol consumption and no other drug (alcohol only) accounted for 5% of drug-related ED visits, accidental ingestion 3%, and malicious poisoning 0.3%.

Drug misuse and abuse in ED visits (Table 3)

Among the nearly 2 million ED visits that were drug-related in 2004, DAWN estimates nearly 1.3 million were associated with drug misuse or abuse. This figure includes 940,953 (CI: 773,124 to 1,108,782) drug-related ED visits that involved illicit drugs or alcohol, and 495,732 (CI: 408,285 to 583,179) ED visits associated with non-medical use of pharmaceuticals.

ED visits involving illicit drugs alone accounted for 30% of all visits related to drug misuse/abuse in 2004. ED visits involving non-medical use of pharmaceuticals alone accounted for another 25%. Only 8% of drug misuse/abuse visits were related to consumption of alcohol by a minor. The remaining visits (37%) involved some combination of illicit drugs, alcohol, and/or pharmaceuticals.

ED visits in each of the three major categories—illicit drugs, alcohol, and non-medical use of pharmaceuticals—are discussed in greater detail in separate sections in the remainder of this publication.

 

Table 2
Drug-related ED visits, by type of case: 2004
Drug-related ED visits
Type of case Unweighted sample data Weighted estimates1 Relative
standard error
(RSE)
95% Confidence interval
Lower
bound
- Upper
bound
Suicide attempt 16,169 121,585 5.3 108,955 - 134,215
Seeking detox 28,800 177,879 30.7 70,845 - 284,913
Alcohol only (age < 21) 11,315 96,809 10.9 76,127 - 117,491
Adverse reaction 71,175 592,044 8.9 488,768 - 695,320
Overmedication 28,707 244,330 10.5 194,046 - 294,614
Malicious poisoning 747 6,026 16.6 4,066 - 7,986
Accidental ingestion 5,796 57,940 7.0 49,990 - 65,890
Other 116,855 701,381 10.6 555,663 - 847,099
Total drug-related visits 279,564 1,997,993 7.4 1,708,205 - 2,287,781
Total ED visits (all reasons) 15,568,029 105,978,433 7.5 90,399,603 - 121,557,263
Drugs
Type of case Unweighted sample data Weighted estimates2 Relative
standard error
(RSE)
95% Confidence interval
Lower bound - Upper bound
Suicide attempt 34,009 266,459 6.1 234,601 - 298,317
Seeking detox 56,272 357,467 32.9 126,957 - 587,977
Alcohol only (age < 21) 11,315 96,809 10.9 76,127 - 117,491
Adverse reaction 92,571 742,916 9.1 610,410 - 875,422
Overmedication 49,893 447,466 11.1 350,115 - 544,817
Malicious poisoning 1,320 10,416 17.0 6,945 - 13,887
Accidental ingestion 7,398 73,992 6.6 64,421 - 83,563
Other 202,018 1,291,276 9.8 1,019,010 - 1,503,542
Drugs in all drug-related visits2 454,796 3,256,802 7.4 2,784,436 - 3,729,168
1 These are estimates of ED visits based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the U.S.
2 These are estimates of drugs. A single ED visit may involve multiple drugs.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

 

Figure 3
Drug-related ED visits, by type of case: 2004

Figure 3   D

SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

 

Table 3
Drug misuse and abuse in ED visits in the U.S., by type of drug involvement: 2004
Drug involvement1 Estimated
visits2,3
Percent
All types of drug misuse/abuse 1,254,078 100%
Illicit drugs only 379,609 30%
Alcohol only (age < 21) 98,174 8%
Pharmaceuticals only 313,125 25%
Combinations    
   Illicit drugs with alcohol4 190,747 15%
   Illicit drugs with pharmaceuticals 99,535 8%
   Alcohol with pharmaceuticals 125,374 10%
   Illicit drugs with alcohol and pharmaceuticals 47,515 4%
1 This classification of drugs is derived from the Multum Lexicon, Copyright 2005, Multum Information Services, Inc. The classification has been modified to meet DAWN's unique requirements (2005). 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 These are estimates of ED visits based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the U.S.
3 Estimates are all expressed in visits.
4 DAWN excludes alcohol-only visits for adults. Alcohol, when present with other drugs, is included for all ages.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

ILLICIT DRUGS IN ED VISITS

The first method for assessing drug abuse in new DAWN focuses on illicit drugs, regardless of case type.

For 2004, DAWN estimates 940,953 (CI: 773,124 to 1,108,782) drug-related ED visits that involved a major substance of abuse (Table 4). This means that nearly half (47%) of all the drug-related ED visits during the year involved alcohol or an illicit drug.

DAWN estimates that cocaine was involved in 383,350 (CI: 284,170 to 482,530) ED visits. In other words, approximately one in five drug-related ED visits (19%) involved cocaine.

Marijuana was involved in 215,665 (CI: 175,930 to 255,400) ED visits. Thus, marijuana may be only slightly less common than cocaine in drug-related ED visits.

Heroin was involved in 162,137 (CI: 122,414 to 201,860) drug-related ED visits or 8% of drug-related ED visits overall. This could, however, be an underestimate. Heroin is an opiate, and some drug screens test for opiates only as a class. About three-quarters (74%) of reports of "opiates" submitted to DAWN for 2004 came from toxicology findings, so some unknown quantity of these may have been heroin. The number of unspecified opiates in drug-related ED visits is estimated at 37,007 (CI: 28,738 to 45,276) visits, or 2% of all drug-related ED visits.

Stimulants, including amphetamines and methamphetamine, were involved in 102,843 (CI: 61,520 to 144,166) ED visits, about 5% of drug-related ED visits overall. Amphetamines and methamphetamine are combined for this analysis because more than 8 out of 10 (86%) amphetamine reports are derived from toxicology findings.7 Since some drug screens test for amphetamines only as a class, an amphetamine-positive result could indicate amphetamine or methamphetamine.

Other illicit drugs appeared at much lower frequencies. For 2004, DAWN estimates:

By design, DAWN excludes illicit drugs from all case types except suicide attempt, seeking detox, malicious poisoning, and other. Also by design, most illicit drug use will be classified in case type other, with most of the remainder in suicide attempts and seeking detox cases (Table 5). For example:

 

Table 4
Illicit drugs and alcohol in drug-related ED visits: 2004
Drug category and selected drugs1 Estimated visits2,3,4 Relative
standard error
(RSE)
95% Confidence interval
Lower
bound
- Upper
bound
Total drug-related ED visits 1,997,993 7.4 1,708,205 - 2,287,781
Major substances of abuse (includes alcohol) 940,953 9.1 773,124 - 1,108,782
   Alcohol 461,809 9.5 375,820 - 547,798
      Alcohol-in-combination 363,641 10.4 289,516 - 437,766
      Alcohol alone (age < 21 only) 98,168 10.9 77,196 - 119,140
   Cocaine 383,350 13.2 284,170 - 482,530
   Heroin 162,137 12.5 122,414 - 201,860
   Marijuana 215,665 9.4 175,930 - 255,400
   Stimulants 102,843 20.5 61,520 - 144,166
      Amphetamines 32,686 15.5 22,757 - 42,615
      Methamphetamine 73,400 22.7 40,742 - 106,058
   MDMA (Ecstasy) 8,621 15.6 5,985 - 11,257
   GHB 2,340 48.3 125 - 4,555
   Flunitrazepam (Rohypnol) 473 49.3 16 - 930
   Ketamine 227 26.5 109 - 345
   LSD 1,953 20.2 1,179 - 2,727
   PCP 8,928 22.9 4,920 - 12,936
   Miscellaneous hallucinogens 3,445 18.4 2,202 - 4,688
   Inhalants 9,275 15.5 6,457 - 12,093
   Combinations not tabulated above (NTA) 1,524 16.5 1,032 - 2,016
1 This classification of drugs is derived from the Multum Lexicon, Copyright 2005, Multum Information Services, Inc. The classification has been modified to meet DAWN's unique requirements (2005). 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 These are estimates of ED visits based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the U.S.
3 Estimates are all expressed in visits. For example, 383,350 visits involved cocaine, and 162,137 visits involved heroin. Visits cannot be summed across drugs because drug-related ED visits often involve multiple drugs (e.g., visits involving both cocaine and heroin would be double counted).
4 Three dots (…) indicate that an estimate with an RSE greater than 50% or an estimate less than 30 has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

 

Table 5
Illicit drugs, by type of case: 2004
Drug category and selected drugs1 All case types Type of case
Suicide attempt Seeking detox Alcohol
only (age
< 21)
Adverse reaction Over-medication Malicious poisoning Accidental ingestion Other
Drug-related ED Visits2,3,4
Total drug-related ED visits 1,997,993 121,585 177,879 96,809 592,044 244,330 6,026 57,940 701,381
Cocaine 383,350 13,940 81,439   995 286,648
Heroin 162,137 2,986 53,088   111 46 105,906
Marijuana 215,665 9,747 27,259   879 148 177,380
Stimulants 102,843 4,218 12,151   345 810 84,926
   Amphetamines 32,686 1,894 1,829   341 532 88 27,861
   Methamphetamine 73,400 2,391 10,518   281 60,042
MDMA (Ecstasy) 8,621 278   7,107
GHB 2,340   231 1,751
Flunitrazepam (Rohypnol) 473  
Ketamine 227   144
LSD 1,953 60   1,784
PCP 8,928 418 410   7,779
Miscellaneous halluicinogens 3,445 90   3,214
Inhalants 9,275 187   1,165 3,338 4,376
Combinations NTA 1,524 222   1,282
Percent of visits
Cocaine 19% 11% 46%   17% 41%
Heroin 8% 2% 30%   2% 0% 15%
Marijuana 11% 8% 15%   15% 0% 25%
Stimulants 5% 3% 7%   0% 13% 12%
   Amphetamines 2% 2% 1%   9% 4%
   Methamphetamine 4% 2% 6%   5% 9%
MDMA (Ecstasy) 0% 0%   1%
GHB 0% 0%   4% 0%
Flunitrazepam (Rohypnol) 0%  
Ketamine 0%   0%
LSD 0% 0%   0% 0%
PCP 0% 0% 0%   1%
Miscellaneous hallucinogens 0% 0%   0%
Inhalants 0% 0%   0% 6% 1%
Combinations NTA 0% 0%   0%
ED Visits per 100,000 population2,3,4
Total drug-related ED visits 680 41 61 113 202 83 2 20 239
Cocaine 131 5 28   0 98
Heroin 55 1 18   0 0 36
Marijuana 73 3 9   0 0 60
Stimulants 35 1 4   0 0 29
   Amphetamines 11 1 1   0 0 0 9
   Methamphetamine 25 1 4   0 20
MDMA (Ecstasy) 3 0   2
GHB 1   0 1
Flunitrazepam (Rohypnol) 0  
Ketamine 0   0
LSD 1 0   1
PCP 3 0 0   3
Miscellaneous halluicinogens 1 0   1
Inhalants 3 0   0 1 1
Combinations NTA 1 0   0
1 This classification of drugs is derived from the Multum Lexicon, Copyright 2005, Multum Information Services, Inc. The classification has been modified to meet DAWN's unique requirements (2005). 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 These are estimates of ED visits based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the U.S.
3 Estimates are all expressed in visits. For example, 383,350 visits involved cocaine, and 162,137 visits involved heroin. Visits cannot be summed across drugs because drug-related ED visits often involve multiple drugs (e.g., visits involving both cocaine and heroin would be double counted).
4 Three dots (…) indicate that an estimate with an RSE greater than 50% or an estimate less than 30 has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

When considered in relation to the population of the U.S., ED visits associated with illicit drugs are relatively infrequent, but vary across the major drugs (Figure 4):

The rates of ED visits involving cocaine, marijuana, heroin, and stimulants did not differ between males and females after taking population size and the margin of error into account (Figure 5). The rates for patients aged 21 to 54 tended to be similar for cocaine and heroin, with lower rates for younger and older patients (Table 6 and Figure 5). For marijuana, the rates were highest for patients aged 18 to 24. For stimulants, the rates were highest for patients aged 18 to 44.

 

Figure 4
Illicit drugs in ED visits: 2004

Figure 4   D

SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

 

Table 6
Illicit drugs, by patient characteristics: 2004
Patient characteristics Selected drugs1
Cocaine Heroin Marijuana Stimulants MDMA (Ecstasy) GHB LSD PCP
Drug-related ED visits2,3,4
Total drug-related ED visits 383,350 162,137 215,665 102,843 8,621 2,340 1,953 8,928
Gender                
   Male 249,942 108,768 141,871 58,700 4,916 1,316 1,786 5,783
   Female 133,296 53,319 73,716 44,138 3,704 167 3,131
   Unknown 112 50 78
Age                
   0-5 years 253
   6-11 years 380
   12-17 years 11,539 1,400 39,035 6,402 1,429 451 806
   18-20 years 18,404 8,801 27,742 10,028 2,374 423 551 853
   21-24 years 34,564 18,256 32,154 15,542 2,179 339 1,543
   25-29 years 49,153 25,037 28,645 18,340 1,357 404 157 1,246
   30-34 years 55,142 22,474 24,716 14,484 611 308 133 1,670
   35-44 years 127,662 44,864 40,639 24,405 513 326 201 1,724
   45-54 years 73,807 34,383 19,389 11,663 895
   55-64 years 10,790 5,933 2,311 1,430
   65 years and older 1,503 653 403 49
   Unknown 518 188 136 35
Race/ethnicity                
   White 145,216 68,297 111,685 60,469 4,108 1,326 4,734
   Black 152,732 41,831 53,955 4,323 2,140 37 268 2,133
   Hispanic 36,888 18,595 18,677 8,904 50 104 861
   Race/ethnicity NTA 4,589 1,607 2,706 1,910 191 34
   Unknown 43,925 31,807 28,642 27,238 1,370 291 233 1,165
ED visits per 100,000 population2,3,4
Total drug-related ED visits 131 55 73 35 3 1 1 3
Gender                
   Male 173 75 98 41 3 1 1 4
   Female 89 36 49 0
Age                
   0-5 years 1
   6-11 years 2
   12-17 years 45 6 154 25 6 2 3
   18-20 years 149 71 225 81 19 3 4 7
   21-24 years 205 108 190 92 13 2 9
   25-29 years 251 128 146 94 7 2 1 6
   30-34 years 269 110 121 71 3 2 1 8
   35-44 years 289 102 92 55 1 1 0 4
   45-54 years 177 83 47 28 2
   55-64 years 37 20 8 5
   65 years and older 4 2 1 0
1 This classification of drugs is derived from the Multum Lexicon, Copyright 2005, Multum Information Services, Inc. The classification has been modified to meet DAWN's unique requirements (2005). 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 These are estimates of ED visits based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the U.S.
3 Estimates are all expressed in visits.
4 Three dots (…) indicate that an estimate with an RSE greater than 50% or an estimate less than 30 has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

 

Figure 5
Illicit drugs, ED visit rates by age and gender: 2004

Figure 5   D

SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

ALCOHOL AND DRUG-RELATED ED VISITS

The second method of assessing drug misuse and abuse in DAWN focuses on alcohol:

For 2004, DAWN estimates 461,809 (CI: 375,820 to 547,798) drug-related ED visits involved alcohol in combination with another drug or alcohol alone in a patient under the age of 21. Thus, nearly a quarter (23%) of all drug-related ED visits involved alcohol in one of these forms (Table 7).

 

Table 7
Alcohol in drug-related ED visits: 2004
Drug category and selected drugs1 Estimated visits2,3 Relative
standard error
(RSE)
95% Confidence interval
Lower
bound
- Upper
bound
Total drug-related ED visits 1,997,993 7.4 1,708,205 - 2,287,781
Alcohol 461,809 9.5 375,820 - 547,798
   Alcohol-in-combination 363,641 10.4 289,516 - 437,766
   Alcohol alone 98,168 10.9 77,196 - 119,140
1 This classification of drugs is derived from the Multum Lexicon, Copyright 2005, Multum Information Services, Inc. The classification has been modified to meet DAWN's unique requirements (2005). 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 These are estimates of ED visits based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the U.S.
3 Estimates are all expressed in visits.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

Alcohol in combination with other drugs (Tables 8-10, Figure 6)

DAWN estimates 363,641 (CI: 289,516 to 437,766) ED visits related to use of alcohol in combination with another drug in 2004. Alcohol is reported to DAWN in combination with other drugs, regardless of the patient's age. These are the only alcohol reports received for patients aged 21 and older. Nearly 9 out of 10 (87%) ED visits implicating alcohol with another drug were for adult patients. Alcohol in combination appeared in substantial numbers in most case types (Table 8):

Alcohol was involved with other drugs in about a quarter (27%) of ED visits involving misuse or abuse of drugs—i.e., overmedication, malicious poisoning, and case type other, considered as a group. Alcohol appeared rarely in adverse reactions (1% of visits).

 

Table 8
Alcohol, by type of case: 2004
Drug category and selected drugs1 All case types Type of case
Suicide attempt Seeking detox Alcohol only (age<21) Adverse reaction Over-medication Malicious poisoning Accidental ingestion Other
Drug-related ED visits2,3,4
Total drug-related ED visits 1,997,993 121,585 177,879 96,809 592,044 244,330 6,026 57,940 701,381
Alcohol 461,809 37,414 60,022 96,809 8,212 47,915 2,935 603 207,897
   Alcohol-in-
   combination
363,641 36,702 59,599   8,200 47,915 2,935 601 207,689
   Alcohol alone 98,168 712 424 96,809
Percent of visits
Alcohol 23% 31% 34% 100% 1% 20% 49% 1% 30%
   Alcohol-in-
   combination
18% 30% 34%   1% 20% 49% 1% 30%
   Alcohol alone 5% 1% 0% 100%
1 This classification of drugs is derived from the Multum Lexicon, Copyright 2005, Multum Information Services, Inc. The classification has been modified to meet DAWN's unique requirements (2005). 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 These are estimates of ED visits based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the U.S.
3 Estimates are all expressed in visits.
4 Three dots (…) indicate that an estimate with an RSE greater than 50% or an estimate less than 30 has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

Alcohol was most frequently combined with (Table 9):

Among cases involving misuse or abuse of drugs, DAWN estimates 258,539 (CI: 189,623 to 327,455) ED visits involving alcohol in combination with other drugs in 2004. Males accounted for 62% of these visits involving alcohol and other drugs, but taking population size into account, males and females had similar rates of such visits. There was little variation in rates across the age groups from ages 18 to 44. However, rates of such visits were lower for older and younger patients.

In terms of race and ethnicity, 51% of the visits with alcohol in combination involved patients who were white. Evaluating the relative frequencies of the other race/ethnicity groups is impeded by missing data; in 14% of visits race/ethnicity was unknown.

 

Figure 6
Alcohol with other drugs, ED visit rates by age and gender: 2004

Figure 6   D

SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

 

Table 9
Drugs reported most frequently with alcohol, by type of case: 2004
Drugs reported with alcohol1 All case types Type of case
Suicide attempt Seeking detox Alcohol
only
(age < 21)
Adverse reaction Over-medication Malicious poisoning Accidental ingestion Other
Drug-related ED visits2,3,4
No other drug 98,168 712 424 96,809
Cocaine only 83,816 1,566 20,234   324 61,686
Marijuana only 33,963 506 2,333   31,099
Cocaine and marijuana only 19,697 437 4,973   94 14,193
Heroin only 14,669 349 4,565   9,751
Cocaine and heroin only 9,992 167 4,181   5,641
Stimulants only 9,525 204 949   87 8,269
Alprazolam only 9,035 1,371 717   253 4,097 2,590
1 This classification of drugs is derived from the Multum Lexicon, Copyright 2005, Multum Information Services, Inc. The classification has been modified to meet DAWN's unique requirements (2005). 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 These are estimates of ED visits based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the U.S.
3 Estimates are all expressed in visits.
4 Three dots (…) indicate that an estimate with an RSE greater than 50% or an estimate less than 30 has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

 

Table 10
Alcohol, by patient characteristics: 2004
Patient characteristics All case types1,2,3,4 Overmedication, malicious
poisoning, and case
type other1,2,3,4
All alcohol Alcohol-in-
combination
Alcohol alone Alcohol-in-combination
Total drug-related ED visits 461,809 363,641 98,168 258,539
Gender        
   Male 281,019 224,217 56,802 161,412
   Female 180,675 139,322 41,353 97,037
   Unknown 114 102 90
Age        
   0-5 years 701 335 366
   6-11 years 283 267
   12-17 years 60,118 19,605 40,512 16,835
   18-20 years 82,583 25,676 56,907 21,004
   21-24 years 37,437 37,436 28,070
   25-29 years 41,592 41,584 29,309
   30-34 years 44,946 44,935 29,931
   35-44 years 106,723 106,720 74,182
   45-54 years 70,440 70,362 47,537
   55-64 years 13,319 13,314 9,006
   65 years and older 3,298 3,289 2,186
   Unknown 369 369 284
Race/ethnicity        
   White 250,706 191,860 58,846 130,865
   Black 94,014 86,541 7,473 63,102
   Hispanic 44,747 32,773 11,974 24,509
   Race/ethnicity NTA 6,727 4,570 2,157 3,613
   Unknown 65,614 47,896 17,718 36,451
1 This classification of drugs is derived from the Multum Lexicon, Copyright 2005, Multum Information Services, Inc. The classification has been modified to meet DAWN's unique requirements (2005). 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 These are estimates of ED visits based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the U.S.
3 Estimates are all expressed in visits.
4 Three dots (…) indicate that an estimate with an RSE greater than 50% or an estimate less than 30 has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

Alcohol only in patients under the age of 21 (Table 11, Figure 7)

DAWN estimates 96,809 (CI: 76,127 to 117,491) ED visits related to use of alcohol by patients who were younger than age 21 in 2004 (Table 4). These numbers increase very little if instances of underage alcohol use in suicide attempts and seeking detox cases are also included (Table 11).

Alcohol was specifically indicated in a diagnosis in about two out of three (68%) alcohol-only visits, with toxic effects (e.g., "intoxication") in slightly fewer (57%) visits. Injuries were diagnosed in 29% of alcohol-only visits, and accidents, involving falls or motor vehicles, were indicated by diagnosis in 7% (Table 11).

Most (85%) of such visits resulted in patients being treated and released, usually to home; another 9% were admitted to inpatient units.

Taking population size into account, the rate of alcohol-only ED visits for ages 18 to 20 (456 visits per 100,000 population) was 2.9 times that for patients aged 12 to 17 (157 per 100,000). The rates for males and females were equivalent.

In terms of race and ethnicity, 60% of the alcohol-only visits involved patients who were white. Evaluating the relative frequencies of the other race/ethnicity groups is impeded by missing data; in 18% of visits race/ethnicity was unknown.

 

Figure 7
Alcohol only (age < 21), ED visit rates by age and gender: 2004

Figure 7   D

SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

 

Table 11
Alcohol only (age < 21), by patient and visit characteristics: 2004
Patient characteristics Estimated
visits1,2
Visit characteristics Estimated
visits1,2
Total drug-related ED visits 96,809    
Gender   Number of drugs involved  
   Male 56,223    Single drug 96,809
   Female 40,573    Multiple drugs  
   Unknown       Alcohol involved 96,809
Age   Disposition  
   0-5 years 366    Treated and released 82,486
   6-11 years 267       Discharged home 71,324
   12-17 years 39,809       Released to police/jail 9,058
   18-20 years 56,367       Referred to detox/treatment 2,103
   21-24 years      Admitted to this hospital 9,025
   25-29 years         ICU/critical care 3,614
   30-34 years         Surgery 247
   35-44 years         Chemical dependency/detox
   45-54 years         Psychiatric unit 800
   55-64 years         Other inpatient unit 4,106
   65 years and older      Other disposition 5,298
   Unknown         Transferred 3,347
Race/ethnicity         Left against medical advice 1,023
   White 58,010       Died
   Black 7,314       Other 317
   Hispanic 11,757       Not documented 519
   Race/ethnicity NTA 2,147    
   Unknown 17,581    
    Selected diagnoses3  
       Drug-related diagnoses  
          Abuse 6,789
          Alcohol 65,742
          Toxic effects 55,412
       Other conditions  
          Altered mental status 7,635
          Injuries 27,689
          Psychiatric conditions 6,430
             Depression 3,353
          Suicide (other than attempt) 2,248
       Miscellaneous  
          Accidents 6,736
             Fall 907
             Motor vehicle 5,786
1 These are estimates of ED visits based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the U.S.
2 Three dots (…) indicate that an estimate with an RSE greater than 50% or an estimate less than 30 has been suppressed.
3 Components do not sum to total because multiple diagnoses may be reported for a single visit.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

Any alcohol in patients under the age of 21 (Table 12)

Alcohol use by minors also occurs in combination with other drugs. Considering alcohol only and alcohol in combination with other drugs, DAWN estimates:

 

Table 12
Alcohol in drug-related ED visits in patients under age 21: 2004
Drug category and selected drugs1 Estimated
visits2,3
Relative
standard error
(RSE)
95% Confidence interval
Lower
bound
- Upper
bound
Patients aged 12-17
Alcohol 60,118 12.9 44,918 - 75,318
   Alcohol-in-combination 19,605 14.1 14,187 - 25,023
   Alcohol alone 40,512 13.5 29,793 - 51,231
Patients aged 18-20
Alcohol 82,583 9.1 67,853 - 97,313
   Alcohol-in-combination 25,676 7.3 22,002 - 29,350
   Alcohol alone 56,907 11.1 44,526 - 69,288
1 This classification of drugs is derived from the Multum Lexicon, Copyright 2005, Multum Information Services, Inc. The classification has been modified to meet DAWN's unique requirements (2005). 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 These are estimates of ED visits based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the U.S.
3 Estimates are all expressed in visits.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

NON-MEDICAL USE OF PHARMACEUTICALS
AND DRUG-RELATED ED VISITS

The third method for assessing drug misuse and abuse in DAWN focuses on the non-medical use of prescription and over-the-counter (OTC) pharmaceuticals. For this assessment, we consider drug-related ED visits related to three case types: overmedication, malicious poisoning, and case type other. Overmedication is defined as a patient taking more than the prescribed or recommended dose of a prescription or OTC pharmaceutical, and illicit drugs are explicitly excluded. Malicious poisoning means the patient was deliberately poisoned or drugged by another person. These cases may include drug-facilitated assault, drug-facilitated sexual assault, homicide when the weapon was a drug, and product tampering. Visits classified as case type other include pharmaceuticals taken, in general, for non-medical purposes that do not meet the criteria for any other DAWN case types, including pharmaceuticals taken in combination with illicit drugs. DAWN tries to capture only drugs related to the ED visit and actively discourages reporting of current medications that are unrelated. It is important to understand, however, that it is not possible to eliminate completely the reporting of current medications from the drugs being misused, and this should be considered in interpreting these findings.

Non-medical use of pharmaceuticals (Tables 13-14, Figure 8)

For 2004, DAWN estimates 495,732 (CI: 408,285 to 583,179) ED visits involved non-medical use—i.e., misuse or abuse—of prescription or OTC pharmaceuticals or dietary supplements (Table 13). Multiple drugs were involved in more than half (57%) of these ED visits (Table 14):

Central nervous system (CNS) agents (53% of visits involving non-medical use) and psychotherapeutic agents (48%) were the most frequent drugs in these visits (Table 13). Respiratory agents (4%), cardiovascular agents (6%), and all other categories of pharmaceuticals were much less frequent.

Among the CNS agents, the most frequent drugs were opiate/opioid analgesics (32% of non-medical use visits), including single-ingredient (e.g., oxycodone) and combination forms (e.g., hydrocodone with acetaminophen). Methadone and single-ingredient and combination forms of oxycodone and hydrocodone were the most frequent opioids. Once the margin of error is taken into account, these three opioids appear in similar numbers of visits:

It is not possible to know the extent to which the source of these drugs is a legitimate prescription versus other sources. For example, it is not possible to distinguish methadone used for treatment of opiate addiction from the methadone in pill form that is prescribed for pain.

The opioids were followed in frequency by the non-opioid analgesics containing acetaminophen (8% of visits), muscle relaxants (6%), and anticonvulsants (5%). DAWN estimates 37,512 (CI: 29,057 to 45,967) ED visits involving non-opioid acetaminophen products. The most frequent muscle relaxant in ED visits was carisoprodol, which was involved in 17,366 (CI: 11,170 to 23,562) or 4% of ED visits in 2004.

Among the psychotherapeutic agents, the anxiolytics (anti-anxiety agents), sedatives, and hypnotics are the most frequent, occurring in more than a third (35%) of visits associated with pharmaceutical misuse/abuse. This category of pharmaceuticals includes barbiturates, benzodiazepines, and CNS stimulants such as methylphenidate. ED visits involving benzodiazepines clearly outnumber those involving any of the other types of psychotherapeutic agents. DAWN estimates that 144,385 (CI: 115,520 to 173,250) ED visits associated with pharmaceutical misuse/abuse involved benzodiazepines in 2004. This is comparable to the number for opiates/opioids.

According to DAWN, alprazolam in 49,842 visits (CI: 31,085 to 68,599) and clonazepam in 26,238 visits (CI: 20,581 to 31,895) are the most frequent benzodiazepines in ED visits related to pharmaceutical misuse/abuse. Benzodiazepines without a specific ingredient named appear in comparable numbers: 37,081 ED visits (CI: 26,470 to 47,692). Benzodiazepines occurring less frequently but still in substantial numbers include diazepam in 15,733 ED visits (CI: 12,064 to 19,402) and lorazepam in 16,926 ED visits (CI: 14,139 to 19,713).

Among the other anxiolytics, sedatives, and hypnotics, the following drugs appear in similar numbers of ED visits:

For the ED visits associated with pharmaceutical misuse/abuse, other psychotherapeutic agents of particular interest include:

Methylphenidate, a CNS stimulant that has recently captured much attention, occurs much less frequently. DAWN estimates 1,541 ED visits (CI: 1,027 to 2,055) associated with pharmaceutical misuse/abuse involved methylphenidate.

 

Table 13
Non-medical use of pharmaceuticals: 2004
Selected drug categories and selected drugs1 Estimated
visits2,3,4
Relative
standard error
(RSE)
95% Confidence interval
Lower
bound
- Upper
bound
Total drug-related ED visits 495,732 9.0 408,285 - 583,179
PSYCHOTHERAPEUTIC AGENTS 239,829 9.2 196,584 - 283,074
   Antidepressants 62,743 9.1 51,551 - 73,935
      MAO inhibitors 86.5 -
      SSRI antidepressants 30,817 12.5 23,267 - 38,367
      Tricyclic antidepressants 10,897 11.0 8,547 - 13,247
      Miscellaneous antidepressants 25,218 9.2 20,671 - 29,765
   Antipsychotics 30,846 13.2 22,865 - 38,827
   Anxiolytics, sedatives, and hypnotics 175,115 9.4 142,851 - 207,379
      Barbiturates 11,064 16.4 7,509 - 14,619
      Benzodiazepines 144,385 10.2 115,520 - 173,250
         Alprazolam 49,842 19.2 31,085 - 68,599
         Clonazepam 26,238 11.0 20,581 - 31,895
         Diazepam 15,733 11.9 12,064 - 19,402
         Lorazepam 16,926 8.4 14,139 - 19,713
         Benzodiazepines-NOS 37,081 14.6 26,470 - 47,692
      Misc. anxiolytics, sedatives, and hypnotics 28,304 8.6 23,533 - 33,075
         Diphenhydramine 9,330 10.6 7,392 - 11,268
         Hydroxyzine 2,468 18.2 1,588 - 3,348
         Zolpidem 11,362 11.1 8,890 - 13,834
         Anxiolytics, sedatives and hypnotics-NOS 2,722 26.5 1,309 - 4,135
      CNS stimulants 7,972 8.6 6,627 - 9,317
         Amphetamine-dextroamphetamine 2,227 16.5 1,508 - 2,946
         Caffeine 2,787 14.3 2,005 - 3,569
         Dextroamphetamine 408 45.1 47 - 769
         Methylphenidate 1,541 17.0 1,027 - 2,055
CENTRAL NERVOUS SYSTEM AGENTS 261,582 8.7 216,976 - 306,188
   Analgesics 222,833 8.1 187,457 - 258,209
      Antimigraine agents 467 23.1 255 - 679
      Cox-2 inhibitors 2,641 18.8 1,667 - 3,615
      Opiates/opioids 158,281 8.7 131,292 - 185,270
         Opiates/opioids, unspecified 29,461 12.4 22,301 - 36,621
      Narcotic analgesics 132,207 9.8 106,813 - 157,601
         Buprenorphine/combinations 236 35.9 69 - 403
         Codeine/combinations 5,836 11.1 4,566 - 7,106
         Fentanyl/combinations 8,000 15.0 5,648 - 10,352
         Hydrocodone/combinations 42,491 12.8 31,831 - 53,151
         Hydromorphone/combinations 2,779 26.1 1,358 - 4,200
         Meperidine/combinations 1,310 22.3 738 - 1,882
         Methadone 31,874 13.0 23,752 - 39,996
         Morphine/combinations 12,558 18.2 8,077 - 17,039
         Oxycodone/combinations 36,559 10.6 28,964 - 44,154
         Propoxyphene/combinations 6,448 16.3 4,388 - 8,508
      Non-steroidal anti-inflammatory agents 22,961 9.7 18,596 - 27,326
         Ibuprofen 17,934 9.4 14,629 - 21,239
         Naproxen 4,817 16.3 3,278 - 6,356
      Salicylates/combinations 11,820 16.1 8,090 - 15,550
      Miscellaneous analgesics/combinations 41,508 10.6 32,884 - 50,132
         Acetaminophen/combinations 37,512 11.5 29,057 - 45,967
         Tramadol 2,984 14.7 2,124 - 3,844
      Analgesic combinations NTA 1,195 19.4 740 - 1,650
   Anorexiants 1,336 20.3 805 - 1,867
   Anticonvulsants 26,926 11.2 21,015 - 32,837
   Antiemetic/antivertigo agents 1,457 24.0 771 - 2,143
   Antiparkinson agents 1,615 18.0 1,045 - 2,185
   General anesthetics 98.8 -
   Muscle relaxants 28,338 15.2 19,896 - 36,780
      Carisoprodol 17,366 18.2 11,170 - 23,562
      Cyclobenzaprine 5,932 14.4 4,258 - 7,606
   Miscellaneous CNS agents 854 28.6 376 - 1,332
RESPIRATORY AGENTS 20,340 11.1 15,914 - 24,766
   Antihistamines 5,148 19.4 3,190 - 7,106
   Bronchodilators 2,351 34.0 785 - 3,917
   Decongestants 1,468 20.4 882 - 2,054
   Expectorants 1,258 23.7 674 - 1,842
   Upper respiratory combinations 9,431 8.9 7,787 - 11,075
   Respiratory agents NTA 1,979 16.1 1,354 - 2,604
CARDIOVASCULAR AGENTS 27,286 17.2 18,088 - 36,484
   Antiadrenergic agents, centrally acting 3,752 25.2 1,898 - 5,606
   Beta-adrenergic blocking agents 7,014 14.8 4,980 - 9,048
   Calcium channel blocking agents 2,465 22.8 1,363 - 3,567
   Diuretics 3,968 30.4 1,604 - 6,332
   Cardiovascular agents NTA 14,886 15.6 10,335 - 19,437
1 This classification of drugs is derived from the Multum Lexicon, Copyright 2005, Multum Information Services, Inc. The classification has been modified to meet DAWN's unique requirements (2005). 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.
2These are estimates of ED visits based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the U.S.
3 Estimates are all expressed in visits. Visits cannot be summed across drugs because drug-related ED visits often involve multiple drugs.
4 Three dots (…) indicate that an estimate with an RSE greater than 50% or an estimate less than 30 has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

 

Figure 8
Non-medical use of pharmaceuticals, ED visit rates by age and gender: 2004

Figure 8   D

SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

Taking population size and the margin of error into account, visits for non-medical use of pharmaceuticals did not differ between females (186 visits per 100,000 population) and males (151 per 100,000 population) (Figure 8). In terms of age, visit rates were highest for patients aged 18 to 44. Visit rates were lowest for patients 11 and younger. In terms of race and ethnicity, 65% of visits involved patients who were white. Evaluating the relative frequencies of the other race/ethnicity groups is impeded by missing data; in 15% of visits race/ethnicity was unknown.

Patients were treated and released in about half (52%) of ED visits associated with non-medical use of pharmaceuticals, but a third (34%) resulted in admission to inpatient hospital units (Table 14). Of those admitted to the hospital, about a third (36%) were sent to a critical care unit, and about 16% of those admitted to inpatient units went to a psychiatric unit. About 9% of ED visits for non-medical use of pharmaceuticals were transferred to another health care facility.

Among the most frequently occurring diagnoses for pharmaceutical misuse were overdose (in 38% of visits), depression or another psychiatric condition (23%), and suicide other than an attempt (11%). Diagnoses classified as other suicide include suicidal gestures, thoughts, or ideation; suicide attempts were classified separately. Visits frequently had diagnoses indicating drug involvement (70%). About 9% of ED visits involving the non-medical use of pharmaceuticals involved a diagnosis of pain.

 

Table 14
Non-medical use of pharmaceuticals, by patient and visit characteristics: 2004
Patient characteristics Estimated
visits1,2
Visit characteristics Estimated
visits1,2
Total drug-related ED visits 495,859    
Gender   Number of drugs involved  
   Male 218,326    Single drug 213,241
   Female 277,273    Multiple drugs 282,618
   Unknown       Alcohol involved 113,136
Age   Disposition  
   0-5 years 3,076    Treated and released 260,169
   6-11 years 1,457       Discharged home 228,902
   12-17 years 46,281       Released to police/jail 10,175
   18-20 years 37,294       Referred to detox/treatment 21,014
   21-24 years 47,210    Admitted to this hospital 168,899
   25-29 years 49,411       ICU/critical care 60,963
   30-34 years 49,461       Surgery 741
   35-44 years 109,938       Chemical dependency/detox
   45-54 years 87,118       Psychiatric unit 27,069
   55-64 years 32,556       Other inpatient unit 75,885
   65 years and older 31,203    Other disposition 66,793
   Unknown 185       Transferred 42,861
Race/ethnicity         Left against medical advice 9,297
   White 322,515       Died 1,434
   Black 58,105       Other 3,533
   Hispanic 36,318       Not documented
   Race/ethnicity NTA 5,315    
   Unknown 73,013    
    Selected diagnoses3  
       Drug-related diagnoses 324,739
          Abuse 80,283
          Drug or Alcohol 360,339
             Alcohol 43,950
             Drug 345,588
          Overdose 190,703
          Overmedication 9,744
          Toxic effects 50,256
       Body system (includes infections) 101,997
       Other conditions  
          Altered mental status 47,406
          Pain 43,499
          Psychiatric conditions 115,383
             Depression 75,634
          Suicide (other than attempt) 54,983
1 These are estimates of ED visits based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the U.S.
2 Three dots (…) indicate that an estimate with an RSE greater than 50% or an estimate less than 30 has been suppressed.
3 Components do not sum to total because multiple diagnoses may be reported for a single visit.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

SPECIAL TYPES OF DRUG-RELATED ED VISITS

This chapter profiles two special types of drug-related ED visits captured by DAWN. Drug-related suicide attempts and seeking detox cases are considered as separate and distinct classes of drug misuse or abuse.

Suicide attempt (Tables 15-16, Figure 9)

DAWN estimates 121,585 (CI: 108,955 to 134,215) ED visits for drug-related suicide attempts in 2004 (Table 15). It is important to remember that DAWN includes only those suicide attempts that involve drugs, but these attempts are not limited to overdoses. Also included are persons who attempt suicide by other means (e.g., by gun) when drugs are involved. Excluded are suicide attempts not involving drugs (e.g., by gun alone) and those documented as something other than an attempt (e.g., suicide ideation, gesture, thought, and so forth).

Nearly two-thirds of ED visits for drug-related suicide attempts involved multiple drugs (64%) (Table 16). Alcohol was the most frequently implicated drug and was involved in nearly a third (31%) of the ED visits for drug-related suicide attempts. Since DAWN excludes visits for adults when alcohol is the only drug, the role of alcohol in suicide attempts is probably larger. The most frequent illicit drugs were cocaine (11% of visits) and marijuana (8% of visits), but the margins of error for the illicit drugs are quite large and the numbers are relatively small when compared with the pharmaceuticals.

More than half (56%) of ED visits for drug-related suicide attempts involved psychotherapeutic agents, and nearly half (47%) involved CNS agents. The most commonly used psychotherapeutic agents were benzodiazepines (26%) and antidepressants (22%), which were implicated in similar numbers of ED visits. Again, it is not possible to know the extent to which these pharmaceuticals may have been prescribed to the patient for a preexisting condition. The CNS agents were primarily analgesics (pain relievers), including both prescription and over-the-counter formulations. DAWN estimates that the most commonly used pain relievers were acetaminophen/combinations and opiates/opioids, with each present in more than a third of visits (39% and 37%, respectively), followed by non-steroidal anti-inflammatory agents (NSAIDS, such as ibuprofen and naproxen, 25%), and salicylates/combinations (aspirins, 11%).

Among the 121,585 ED visits involving suicide, the involvement of alcohol or a drug was indicated by diagnosis in 65% of visits. The other most frequent diagnoses indicated overdose (61%) and suicide attempt (56%). Psychiatric conditions were also implicated in a large proportion of visits (40%), and the psychiatric disorder most frequently present was depression (35%).

About half (54%) of the suicide attempts were admitted for inpatient hospital care, and nearly half of these were admitted to a critical care unit. Others were admitted to psychiatric (14%) or other inpatient units (15%). Another 26% were transferred to another health care facility; only 12% were discharged home. Very few (0.25%) died in the ED. However, DAWN does not account for patients who die before arriving at the ED or patients who die after admission to inpatient units of the hospital.

 

Table 15
Suicide attempt: 2004
Drug category and selected drugs1 Estimated visits2,3,4 Relative
standard error
(RSE)
95% Confidence interval
Lower
bound
- Upper
bound
Total drug-related ED visits 121,585 5.3 108,955 - 134,215
Major substances of abuse
Alcohol 37,414 7.2 32,134 - 42,694
   Alcohol-in-combination 36,702 7.2 31,522 - 41,882
   Alcohol alone 712 27.4 330 - 1,094
Cocaine 13,940 12.0 10,661 - 17,219
Heroin 2,986 21.0 1,757 - 4,215
Marijuana 9,747 12.6 7,340 - 12,154
Stimulants 4,218 21.0 2,481 - 5,955
   Amphetamines 1,894 22.4 1,063 - 2,725
   Methamphetamine 2,391 25.8 1,182 - 3,600
MDMA (Ecstasy) 278 36.2 80 - 476
GHB 89.4 -
Flunitrazepam (Rohypnol) 92.9 -
Ketamine 43.7 -
LSD 59.5 -
PCP 418 26.1 204 - 632
Miscellaneous hallucinogens 91.9 -
Inhalants 187 38.4 46 - 328
Combinations NTA 58.8 -
Other substances
PSYCHOTHERAPEUTIC AGENTS 68,238 5.6 60,749 - 75,727
   Antidepressants 26,787 6.7 23,269 - 30,305
      MAO inhibitors 0.0 -
      SSRI antidepressants 13,968 7.7 11,859 - 16,077
      Tricyclic antidepressants 2,561 15.3 1,793 - 3,329
      Miscellaneous antidepressants 12,150 7.9 10,268 - 14,032
   Antipsychotics 12,830 9.3 10,492 - 15,168
   Anxiolytics, sedatives, and hypnotics 42,967 6.5 37,493 - 48,441
      Barbiturates 1,004 19.7 616 - 1,392
      Benzodiazepines 31,695 8.0 26,724 - 36,666
         Alprazolam 11,451 12.6 8,623 - 14,279
         Clonazepam 8,370 11.2 6,533 - 10,207
         Diazepam 3,571 14.4 2,564 - 4,578
         Lorazepam 4,973 13.3 3,677 - 6,269
         Benzodiazepines-NOS 3,619 19.4 2,243 - 4,995
         Misc. anxiolytics, sedatives, and hypnotics 12,988 8.1 10,926 - 15,050
            Diphenhydramine 4,718 13.8 3,442 - 5,994
            Hydroxyzine 1,672 15.2 1,174 - 2,170
            Zolpidem 4,408 11.0 3,457 - 5,359
            Anxiolytics, sedatives and hypnotics-NOS 1,140 20.1 691 - 1,589
         CNS stimulants 1,457 19.5 900 - 2,014
            Amphetamine-dextroamphetamine 289 39.5 66 - 512
            Caffeine 51.8 -
            Dextroamphetamine 80.4 -
            Methylphenidate 348 33.5 119 - 577
CENTRAL NERVOUS SYSTEM AGENTS 56,763 6.3 49,754 - 63,772
   Analgesics 46,259 6.0 40,818 - 51,700
      Antimigraine agents 299 39.3 68 - 530
      Cox-2 inhibitors 708 19.7 436 - 980
      Opiates/opioids 16,889 7.9 14,274 - 19,504
         Opiates/opioids, unspecified 1,874 14.9 1,327 - 2,421
      Narcotic analgesics 15,133 8.6 12,583 - 17,683
         Buprenorphine/combinations 101.8 -
         Codeine/combinations 1,431 16.4 970 - 1,892
         Fentanyl/combinations 65.3 -
         Hydrocodone/combinations 7,325 11.0 5,745 - 8,905
         Hydromorphone/combinations 54.1 -
         Meperidine/combinations 57.7 -
         Methadone 1,207 28.4 535 - 1,879
         Morphine/combinations 683 23.8 364 - 1,002
         Oxycodone/combinations 3,324 12.1 2,536 - 4,112
         Propoxyphene/combinations 2,088 19.3 1,298 - 2,878
      Non-steroidal anti-inflammatory agents 11,594 7.7 9,844 - 13,344
         Ibuprofen 8,063 9.1 6,624 - 9,502
         Naproxen 3,199 16.0 2,195 - 4,203
      Salicylates/combinations 5,068 12.4 3,837 - 6,299
      Miscellaneous analgesics/combinations 19,019 7.9 16,073 - 21,965
         Acetaminophen/combinations 17,847 8.3 14,944 - 20,750
         Tramadol 1,045 27.5 482 - 1,608
      Analgesic combinations NTA 428 32.9 152 - 704
      Anorexiants 115 33.8 39 - 191
      Anticonvulsants 8,643 15.1 6,085 - 11,201
      Antiemetic/antivertigo agents 54.7 -
      Antiparkison agents 246 41.5 46 - 446
      General anesthetics 0.0 -
      Muscle relaxants 5,829 13.5 4,286 - 7,372
         Carisoprodol 2,489 19.1 1,558 - 3,420
         Cyclobenzaprine 1,996 16.7 1,343 - 2,649
      Miscellaneous CNS agents 64.9 -
RESPIRATORY AGENTS 5,879 12.6 4,427 - 7,331
   Antihistamines 1,384 26.4 669 - 2,099
   Bronchodilators 400 37.9 102 - 698
   Decongestants 429 31.2 166 - 692
   Expectorants 347 33.7 118 - 576
   Upper respiratory combinations 3,098 13.7 2,267 - 3,929
   Respiratory agents NTA 625 36.9 172 - 1,078
CARDIOVASCULAR AGENTS 6,258 13.1 4,651 - 7,865
   Antiadrenergic agents, centrally acting 592 22.6 329 - 855
   Beta-adrenergic blocking agents 2,205 17.6 1,445 - 2,965
   Calcium channel blocking agents 766 28.4 339 - 1,193
   Diuretics 459 29.3 196 - 722
   Cardiovascular agents NTA 3,007 18.1 1,941 - 4,073
1 This classification of drugs is derived from the Multum Lexicon, Copyright 2005, Multum Information Services, Inc. The classification has been modified to meet DAWN's unique requirements (2005). 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 These are estimates of ED visits based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the U.S.
3 Estimates are all expressed in visits. Visits cannot be summed across drugs because drug-related ED visits often involve multiple drugs.
4 Three dots (…) indicate that an estimate with an RSE greater than 50% or an estimate less than 30 has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

 

Table 16
Suicide attempt, by patient and visit characteristics: 2004
Patient characteristics Estimated
visits1,2
Visit characteristics Estimated
visits1,2
Total drug-related ED visits 121,585    
Gender   Number of drugs involved  
   Male 45,091    Single drug 43,425
   Female 76,475    Multiple drugs 78,160
   Unknown       Alcohol involved 37,414
Age   Disposition  
   0-5 years    Treated and released 20,770
   6-11 years 31       Discharged home 14,589
   12-17 years 15,299       Released to police/jail 1,025
   18-20 years 11,145       Referred to detox/treatment 5,156
   21-24 years 13,180    Admitted to this hospital 65,129
   25-29 years 14,392       ICU/critical care 29,261
   30-34 years 15,685       Surgery 30
   35-44 years 26,041       Chemical dependency/detox
   45-54 years 19,069       Psychiatric unit 16,982
   55-64 years 4,663       Other inpatient unit 18,551
   65 years and older 1,905    Other disposition 35,687
   Unknown       Transferred 31,822
Race/ethnicity         Left against medical advice 506
   White 75,019       Died 308
   Black 14,155       Other 696
   Hispanic 13,572       Not documented
   Race/ethnicity NTA 1,519    
   Unknown 17,319    
    Selected diagnoses3  
       Drug-related diagnoses 82,254
          Drug or alcohol 79,423
             Alcohol 11,897
             Drug 76,036
                Illicits 5,715
                Other or unspecified drug 72,387
             Overdose 73,596
             Toxic effects 12,995
          Other conditions  
             Altered mental status 4,347
             Psychiatric conditions 48,947
                Depression 42,898
          Suicide 91,561
             Suicide attempt 68,123
             Other suicide-related 24,974
1 These are estimates of ED visits based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the U.S.
2 Three dots (…) indicate that an estimate with an RSE greater than 50% or an estimate less than 30 has been suppressed.
3 Components do not sum to total because multiple diagnoses may be reported for a single visit.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

After accounting for population size and the margins of error, the rate of suicide visits for females (51 visits per 100,000 population) was higher than that for males (31 per 100,000). The rates for patients aged 18 to 34 exceeded the rates for younger and older age groups. Although the rate for patients aged 12 to 17 (60 visits per 100,000) was lower than that for ages 18 to 34, it exceeded the rate for patients aged 45 and over. Nearly two-thirds (62%) of the suicide attempts involved patients who were white. Evaluating the relative frequencies of the other race/ethnicity groups is impeded by missing data; in 14% of visits race/ethnicity was unknown.

 

Figure 9
Suicide attempt, ED visit rates by age and gender: 2004

Figure 9   D

SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

Seeking detox (Tables 17-18, Figure 10)

DAWN estimates 177,879 (CI: 70,845 to 284,913) drug-related ED visits for patients seeking detoxification or substance abuse treatment services during 2004. These visits tend to be concentrated in hospitals with administrative practices that require medical clearance in the ED for admission to detox or substance abuse treatment units in the hospital. Therefore, it is impossible to know the full extent of the demand for these services from this estimate.

More than 60% of the seeking detox ED visits involved multiple drugs. A third (34%) involved alcohol, but for adults this includes only alcohol in combination with other drugs. Among the other major substances of abuse, cocaine (in 46% of visits) and heroin (30% of visits) occurred most frequently, followed by marijuana (15% of visits) and amphetamine or methamphetamine stimulants (7% of visits). Estimates for most pharmaceuticals are too imprecise for publication.9

Drug and/or alcohol involvement was indicated by diagnosis in 9 out of every 10 seeking detox visits. Psychiatric conditions were diagnosed in 23% of the visits. This suggests co-occurring drug misuse/abuse and mental health disorders. Drug abuse was indicated by diagnosis in 57% of the visits.

Among the seeking detox ED visits, 7 out of 10 received some type of follow-up care, either inpatient admission, referral elsewhere for detox or substance abuse treatment services, or transfer to another health care facility. However, about a quarter of seeking detox cases may not have received the care they sought because they were discharged to home.

 

Table 17
Seeking detox: 2004
Drug category and selected drugs1 Estimated
visits2,3
Relative
standard error
(RSE)
95% Confidence interval
Lower
bound
- Upper
bound
Total drug-related ED visits 177,879 30.7 70,845 - 284,913
Major substances of abuse
Alcohol 60,022 26.4 28,964 - 91,080
   Alcohol-in-combination 59,599 26.6 28,527 - 90,671
   Alcohol alone 424 29.6 177 - 671
Cocaine 81,439 28.6 35,787 - 127,091
Heroin 53,088 19.9 32,381 - 73,795
Marijuana 27,259 22.7 15,131 - 39,387
Stimulants 12,151 33.5 4,172 - 20,130
   Amphetamines 1,829 42.0 324 - 3,334
   Methamphetamine 10,518 33.8 3,550 - 17,486
MDMA (Ecstasy) 50.9 -
GHB 86.5 -
Flunitrazepam (Rohypnol) 0.0 -
Ketamine 15.9 -
LSD 60 21.6 35 - 85
PCP 410 38.7 98 - 722
Miscellaneous hallucinogens 90 43.0 14 - 166
Inhalants 52.7 -
Combinations NTA 222 41.3 42 - 402
1 This classification of drugs is derived from the Multum Lexicon, Copyright 2005, Multum Information Services, Inc. The classification has been modified to meet DAWN's unique requirements (2005). 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 These are estimates of ED visits based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the U.S.
3 Estimates are all expressed in visits. Visits cannot be summed across drugs because drug-related ED visits often involve multiple drugs.
4 Three dots (…) indicate that an estimate with an RSE greater than 50% or an estimate less than 30 has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

 

Table 18
Seeking detox, by patient and visit characteristics: 2004
Patient characteristics Estimated
visits1,2
Visit characteristics Estimated
visits1,2
Total drug-related ED visits 177,879    
Gender   Number of drugs involved  
   Male 115,316    Single drug 66,182
   Female 62,541    Multiple drugs 111,696
   Unknown       Alcohol involved 60,022
Age   Disposition  
   0-5 years    Treated and released 89,851
   6-11 years       Discharged home 45,459
   12-17 years 3,000       Released to police/jail 1,174
   18-20 years 10,036       Referred to detox/treatment 43,217
   21-24 years 21,573    Admitted to this hospital 71,017
   25-29 years 28,827       ICU/critical care 450
   30-34 years 28,335       Surgery
   35-44 years 50,380       Chemical dependency/detox
   45-54 years 30,434       Psychiatric unit 11,514
   55-64 years 4,566       Other inpatient unit 4,401
   65 years and older 579    Other disposition 17,011
   Unknown       Transferred 10,433
Race/ethnicity         Left against medical advice 2,539
   White 110,518       Died
   Black 41,128       Other 1,428
   Hispanic 9,641       Not documented 2,532
   Race/ethnicity NTA 1,324    
   Unknown 15,268    
    Selected diagnoses3  
       Drug-related diagnoses 158,636
          Abuse 101,454
          Addiction 14,056
          Dependence 28,052
          Detox 15,505
          Withdrawal 17,179
       Drug or alcohol 161,601
          Alcohol 25,475
          Drug 157,866
             Illicits 47,469
             Other or unspecified drug 117,835
          Other conditions  
             Psychiatric conditions 40,520
             Suicide (other than attempt) 11,470
1 These are estimates of ED visits based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the coterminous U.S.
2 Three dots (…) indicate that an estimate with an RSE greater than 50% has been suppressed.
3 Components do not sum to total because multiple complaints or multiple diagnoses may be reported for a single visit.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2003 (03/2004 update).

Taking population size and the margins of error into account, the rate of seeking detox cases was similar across all age groups in the 18 to 54 range. The rate of seeking detox visits for males was not significantly different than that for females. The lack of significant differences between age and gender subgroups is partially due to large margins of error. The majority (62%) of seeking detox visits involved patients who were white. Evaluating the relative frequencies of the other race/ethnicity groups is impeded by missing data; in 9% of visits race/ethnicity was unknown.

 

Figure 10
Seeking detox, ED visit rates by age and gender: 2004

Figure 10   D

SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

 

Appendixes

 

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 END USER'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 © 2005

Cerner Multum, Inc.
Colorado Center Tower One
2000 South Colorado Boulevard
Suite 11000
Denver, Colorado 80222
Document revised November 30, 2000

Downloaded 12/7/2005 from: http://www.multum.com/license.htm
Lexicon Registration

APPENDIX B
DAWN METHODOLOGY

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 (USDHHS), where the National Institute on Drug Abuse conducted DAWN from 1980 to 1992. Since 1992, the Office of Applied Studies (OAS) of the Substance Abuse and Mental Health Services Administration (SAMHSA), USDHHS, 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 the National Institute on Drug Abuse 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 coterminous 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. Hospitals eligible for selection into the DAWN sample must be non-Federal, short-stay, general surgical and medical hospitals located in the United States with at least one 24-hour emergency department. This current approach was implemented in the 2004 data collection year, and this publication is the first to include estimates based on this sample design.

Under the current methodological design, medical charts for all ED visits within the selected hospitals are reviewed retrospectively to find the drug-related cases for submission to DAWN. DAWN includes ED visits associated with substance abuse and drug 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, a drug needs only to be implicated in the visit; the drug does not need to have caused the visit. Only recent drug use is included; the reason a patient used the drug is irrelevant, and the case criteria are broad enough to encompass all types of drug-related events, which include, but are not limited to, explicit drug abuse. This approach, which finds ED visits related to drug abuse only indirectly, recognizes that medical records (the source of DAWN data) frequently lack explicit documentation of substance abuse, and distinctions between use, misuse, and abuse of drugs are often subjective.

DAWN uses the data from the visits classified as DAWN cases in the selected hospitals to calculate various estimates of drug-related visits for the Nation as a whole, as well as for specific metropolitan areas. To calculate these estimates and measure their precision, the DAWN survey requires the application of sampling and weighting methodologies. This appendix documents the sampling, weighting, and variance estimation methodologies used to develop estimates based on data collected in 2004.

Target population

The target population is drug-related emergency department visits in non-Federal, short-stay, general surgical and medical hospitals in the United States with at least one 24-hour emergency department.

Sampling 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%.10 A health care organization must meet several criteria to be classified as a hospital. These 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 with at least one 24-hour emergency department. Many DAWN hospitals operate multiple emergency departments.

Sample maintenance

DAWN is a longitudinal survey that will be used to analyze trends in drug-related ED visits. In order to keep the frame representative of the current population of hospitals, annual sample updates must be performed. 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 based on updates to the AHA files. These updates include newly eligible hospitals, which are those 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 based on the sampling fraction of the stratum in which the newly eligible hospital is located.

Determination of DAWN eligibility

A hospital is considered ineligible if any one of the key criteria that define eligibility is not met. Only those hospitals that meet all the criteria are considered eligible. For hospitals where critical eligibility data are missing from the AHA database, if one of the non-missing criteria is not met, the hospital is considered ineligible. Otherwise, the hospital is considered to have unknown eligibility. For any hospital with unknown eligibility, other variables on the AHA Annual Survey Database are used to determine eligibility. If the hospital's eligibility remains unknown after exploration of these additional characteristics, then the hospital may be contacted directly to determine eligibility.

Stratification

DAWN employs a stratified simple random sampling approach to select a representative sample of hospitals for inclusion in the DAWN sample. It is important that DAWN produce reliable estimates for major metropolitan areas as well as the Nation. Therefore, the first level of stratification is based on geography. There are two geographic stratification schemes: one for specified Metropolitan Statistical Areas11 and subdivisions, and one for the remainder of the Nation. The second level of stratification is based on ownership and hospital size.

Metropolitan Statistical Areas and subdivisions. In order 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.12 Retention of the existing 21 metropolitan areas was important because there was significant demand 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 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, there continued to be strong interest among users of DAWN statistics in the areas covered by the original 21 metropolitan areas. In order to address the needs of these users, four of the merged areas were subdivided. For each of these areas, there was a sample for the metropolitan area, as well as a sample for each subdivision. This would enable DAWN to produce estimates for the metropolitan areas and for the subdivisions. As a result of this process, the final metropolitan-area sample included a total of 53 geographic units: 48 metropolitan areas, two subdivisions each for three of these metropolitan areas, and three subdivisions for one of these metropolitan areas.

This design recognized that, although each of the 53 geographic units was sampled, not every geographic unit would be active in DAWN at any particular point in time.13 One more feature of the design was needed to preserve this flexibility. When any geographic unit was inactive, it had to be represented in the national estimate and, consequently, in the supplemental sample. Therefore, within each metropolitan area, hospitals were also sampled to serve as that metropolitan area's contribution to the supplemental sample.

Supplemental sample. The sample for the remainder of the Nation is referred to as the supplemental sample because it is designed to supplement the samples from the metropolitan areas to yield a national sample. The supplemental sample is, in effect, the 54th geographic unit for DAWN and is essential to achieve full coverage of the United States. The supplemental sample was formed by first dividing the United States into four Census regions. At any point in time, the supplemental sample provides coverage for all areas outside of the 53 metropolitan units described above, plus sample representation for the metropolitan areas where DAWN is not active.

Stratification by ownership and size. Within the geographic stratification scheme described above, hospitals were further stratified by ownership (public or private) and by size (based on the total number of emergency department visits reported for the hospital in the AHA Annual Survey Database). To begin, a cross classification was created by categories of ownership status and geographic unit. Within each combination of geographic area and ownership status, 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 supplemental 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 supplemental sample.14

Target levels of precision

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% for metropolitan area estimates and RSEs less than or equal to 15% 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 when setting sample size targets.

Sample size and sample allocation

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 WhSh / Ch, 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 the above considerations, sampling rates (i.e., the number of sampled hospitals divided by the number of eligible hospitals) were also subject to the following constraints:

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. There are additional sources of error, often referred to as bias, that 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% coverage of the target population. To minimize possible 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 detail on the survey methodologies used to enhance DAWN data quality and reduce bias are provided in the DAWN 2003 interim estimates of drug-related ED visits.15

Sampling weights

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

Weighting adjustment for nonresponse

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

When the nonresponse adjustment factors were considered final, a nonresponse-adjusted sampling weight was then calculated. For responding hospitals, the nonresponse-adjusted sampling weight was calculated as the product of the nonresponse adjustment factor and the sampling weight. For nonresponding hospitals, the hospital nonresponse-adjusted sampling weight was set to zero. 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 based on 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 to 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 poststratified 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 weighted total visits was equal to the corresponding AHA count of total visits from each stratum.

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 summing the poststratified weights for all data records in question.

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 relative standard error (RSE), which is defined as the SE expressed as a percentage of the value of the estimate. 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 there are 10,000 estimated visits involving a given drug, and this estimate has an SE of 500 visits, then the RSE value is 5%:

RSE = SE/Estimate
RSE = 500/10,000
RSE = 0.05, or 5%.

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% CI is calculated as:

CI = Estimate ± (1.96 x RSE x Estimate)

where 1.96 comes from the table of normal distribution z-values. Ninety-five percent of the normal distribution lies between the z-values of ±1.96.

Applying the formula to the example above, the 95% 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% Confidence interval: 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 value would fall within the confidence interval 95% of the time.

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 age 18 to 20 in the U.S. is much lower than the number of individuals age 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.

To take the size of the underlying population into account, rates of ED visits or drugs per 100,000 population were calculated using population data from the U.S. Bureau of the Census.16

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.

Population estimates used to generate rates for this publication are provided in Appendix D.

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 E describes the race and ethnicity data reported for DAWN.

Publication criteria

DAWN can produce estimates for thousands of patient characteristics, visit characteristics, and drugs. However, some of these estimates are too imprecise or too small to be reliable. In these situations, the estimate was replaced by three dots (…) in the published table. Estimates were suppressed (i.e., not published) according to the following rules.

When the RSE is greater than 50%, the lower bound of the 95% confidence interval approaches or includes the value zero. A confidence interval that includes zero means that the estimate is not statistically different from zero at this precision level.

Estimates this small constitute rare events, which are based on a small number of cases and have precision levels that are difficult to quantify. In many instances, such rare events have variances so large that the estimate would be suppressed based on its RSE alone. Rare events that meet RSE criteria for publication are nonetheless based on very little data and are deemed too unreliable for publication.

There are some estimates with an RSE equal to zero. This occurs when the number of ED visits being estimated is small and all the hospitals contributing to that estimate were selected with certainty, that is, their sampling probability is unity. Strictly speaking, there is no sampling error in such situations and the RSE is equal to zero. These results occur almost exclusively in situations with small numbers of ED visits, where the absence of any sampled hospital data is due to nonresponse and the small number of hospitals contributing to the estimates. In these situations, the necessary data are not available to approximate sampling errors.

APPENDIX C
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 (OTC) 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 exceptions that are excluded from this category.

Alcohol only (age less than 21):
This category of drug-related ED visits includes those in which alcohol was the only drug involved and the patient was aged less than 21. Although alcohol is an illegal drug for minors, combining these cases with other cases involving illicit drugs tends to mask rather than highlight their importance for prevention and treatment efforts.

Case description:
A description of how the drug(s) was related to the patient's ED visit. The case description, in conjunction with the presenting chief complaint and diagnoses, 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:
A "confidence interval" (CI) is an interval estimate, that is, a range of values around a point estimate that takes sampling error into account. Ninety-five percent is an accepted standard of confidence. Technically, a 95% 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% of the time. Practically, a 95% CI summarizes both the estimate and its margin of error in a straightforward way with a reasonable degree of confidence. Calculation of 95% CIs is discussed in Appendix B.

Diagnosis:
The condition(s) for which the patient was treated as determined by the clinician after study. As many as four diagnoses can be entered for each DAWN case.

Disposition:
The location or facility to which an ED patient was referred, transferred, or released. Treated and released includes three categories:
  • Discharged home—"Home" is used as a broad category to mean discharged to the patient's residence. Home is generally used for people who live locally; however, for students at nearby universities, home means their university; for travelers who get sick on the road, it may mean their hotel or wherever they are staying, and so forth.
  • Released to police/jail
  • Referred to detox/treatment—The chart indicates that the patient was referred to a substance abuse treatment or detox program, facility, or provider.

Admitted to this hospital includes five categories of inpatient units:
  • ICU/critical care
  • Surgery
  • Chemical dependency/detox
  • Psychiatric unit
  • Other inpatient unit—The inpatient unit was not specified or does not match one of the preceding units.

Other Disposition includes five categories:
  • Transferred—The patient was transferred to another health care facility.
  • Left against medical advice—The patient left the treatment setting without a physician's approval.
  • Died—The patient died after arriving in the ED but before being discharged, admitted, or transferred.
  • Other—The discharge status is documented in the chart but does not fit into any of the preceding categories.
  • Not documented—The discharge status was not documented in the medical chart.

Drug:
This refers to a substance that was recorded in a DAWN case report. Substances accepted by DAWN include alcohol, illicit drugs, prescription and over-the-counter pharmaceuticals, dietary supplements, and non-pharmaceutical inhalants. Multiple substances ("drugs") can be reported for each DAWN case. Therefore, the total number of drugs exceeds the total number of DAWN cases reported. (See also Single-drug case.)

Drug category:
A generic grouping of pharmaceuticals and other substances reported to DAWN, based on the classification of Multum Information Services. Multum Information Services is a subsidiary of the Cerner Corporation and a developer of clinical drug information systems and a drug knowledge base. More information is available at http://www.multum.com. In general, the Multum categories follow the therapeutic uses for prescription and over-the-counter pharmaceuticals.

Additional clarification is provided for the following drug categories:
  • Alcohol alone—DAWN collects data on alcohol when used alone only if the patient is under age 21.
  • Alcohol-in-combination—Alcohol-in-combination is the category for alcohol present with another reportable substance. DAWN does not gather data on alcohol used alone if the patient is over age 21. For patients 21 and older, alcohol must be used with another substance to be reported to DAWN. Alcohol-in-combination is reportable for all ages.
  • Amphetamines—This class of substances has been extracted from the category of central nervous system (CNS) stimulants because of its importance as a major substance of abuse. For purposes of classification, "amphetamines" (plural) includes a class of compounds derived from or related to the drug amphetamine. Although some "designer" drugs fall into the class of amphetamines, we choose to report some of them individually as major substances of abuse (e.g., methamphetamine). This category does not include other CNS stimulants, such as caffeine or methylphenidate.
  • Combinations not tabulated above (NTA)—This category includes combinations composed of two or more major substances of abuse that are mixed and taken together. For example, "speedball," which usually refers to the combination of heroin and cocaine taken at once, would be classified as a combination NTA, whereas heroin and cocaine used separately would be classified separately in the categories heroin and cocaine. Combinations consisting of a major substance of abuse and another substance are classified in the category of the major substance (e.g., heroin with scopolamine is classified as heroin).
  • Inhalants—This category includes anesthetic gases and psychoactive non-pharmaceutical substances for which the documented route of administration was inhaled, sniffed, or snorted. Psychoactive non-pharmaceuticals fall into one of the following three categories: (1) volatile solvents—adhesives (model airplane glue, rubber cement, household glue), aerosols (spray paint, hairspray, air freshener, deodorant, fabric protector), solvents and gases (nail polish remover, paint thinner, correction fluid and thinner, toxic markers, pure toluene, cigar lighter fluid, gasoline, carburetor cleaner, octane booster), cleaning agents (dry cleaning fluid, spot remover, degreaser), food products (vegetable cooking spray, dessert topping spray such as whipped cream, whippets), and gases (butane, propane, helium); (2) nitrites—amyl nitrites ("poppers," "snappers") and butyl nitrites ("rush," "locker room," "bolt," "climax," "video head cleaner"); or (3) chlorofluorohydrocarbons (freons). Anesthetic gases (e.g., nitrous oxide, ether, chloroform) are presumed to have been inhaled.

Drug misuse and abuse:
A group defined broadly to include case types related to drug misuse or abuse. Additional clarification is provided for the following case types:
  • Overmedication—This category was designed to capture non-medical use, overuse, and misuse of prescription and OTC medications that are not documented as drug abuse in the medical chart.
  • Malicious poisoning—This category was designed to cases of drug use in which the patient was administered a drug by another person for a malicious purpose. Drug-facilitated sexual assault is one type of malicious poisoning, but other types of malicious poisonings such as product tampering would be classified in this category as well.
  • Case type Other—This category includes all drug-related ED visits that could not be assigned to any of the other seven types. By design, most cases of documented drug abuse will fall into this category, and most cases in this category will involve use of illicit drugs or non-medical use of drugs and other substances.

Drug-related ED visit:
Any ED visit related to recent drug use. This is the definition of a DAWN case effective January 1, 2003. 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. It is impossible to determine the number of unique patients involved in the reported drug-related ED visits because no direct patient identifiers are collected by DAWN.

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.

Hospital emergency department (ED):
The unit of a hospital established and staffed to provide emergency medical services. To be eligible for DAWN, the hospital must operate its ED(s) 24 hours a day, 7 days a week.

Malicious poisoning:
See Drug misuse and abuse.

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 nucleus. This DAWN publication utilizes areas defined by the Office of Management and Budget (OMB) in 2003, based on population data from the 2000 decennial Census.

Not otherwise specified (NOS):
Catch-all category for substances that are not specifically named. Terms are classified into an NOS category only when assignment to a more specific category is not possible based on information in the source documentation (ED patient charts).

Not tabulated above (NTA):
Designation used when categories are not presented in complete detail; smaller units are combined in the NTA category.

Overmedication:
See Drug misuse and abuse.

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% 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% are regarded as too imprecise to be published. ED table cells where such estimates would have appeared contain the symbol "…" (3 dots). (See also Relative standard error.)

Race/ethnicity:
The Office of Management and Budget (OMB) is responsible for standard protocols for the collection of data on race and ethnicity by federal systems, including DAWN. In October 1997, OMB issued a revised standard protocol, which permitted separate reporting of race and Hispanic ethnicity, the ability to capture more than one race for an individual, 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. For data collections, such as DAWN, where self-identification of the individual is not feasible, the revised 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.

Since January 2003, DAWN has collected data on race/ethnicity using the combined format. The race/ethnicity categories on the DAWN data collection forms are as follows:
  • White—A person having origins in any of the original peoples of Europe, the Middle East, or North Africa.
  • Black or African American—A person having origins in any of the black racial groups of Africa.
  • Hispanic or Latino—A person of Cuban, Mexican, Puerto Rican, South or Central American, or other Spanish culture or origin, regardless of race.
  • Asian—A person having origins in any of the original peoples of the Far East, Southeast Asia, or the Indian subcontinent including, for example, Cambodia, China, India, Japan, Korea, Malaysia, Pakistan, the Philippine Islands, Thailand, and Vietnam.
  • American Indian or Alaska Native—A person having origins in any of the original peoples of North and South America (including Central America), and who maintains tribal affiliation or community attachment.
  • Native Hawaiian or Other Pacific Islander—A person having origins in any of the original peoples of Hawaii, Guam, Samoa, or other Pacific Islands.
  • Not documented—Used when documentation of race is not available from source records.

Despite the increased detail allowed by these categories and the provision for multiple entries, the actual race/ethnicity data reported to DAWN is quite limited because race and ethnicity are often not documented with this level of specificity in patient/decedent records. As a result, the classification used to tabulate DAWN data has a more limited set of categories, as follows:

  • White—Anyone meeting the definition of white (above). Those who are identified as white and Hispanic are classified as Hispanic.
  • Black—Anyone meeting the definition of black or African American (above). Those who are identified as black or African American and Hispanic are classified as Hispanic.
  • Hispanic—Anyone whose ethnicity is Hispanic or Latino (above) is placed in the category Hispanic, regardless of race.
  • Race/ethnicity NTA—This includes those categories that are too small to report independently including: 2 or more races, American Indian or Alaska Native, Asian, Native Hawaiian or Other Pacific Islander.
  • Unknown—Race/ethnicity is unknown. Those who are identified only as Hispanic are classified as Hispanic.

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 10%. The larger the RSE, the less precise the estimate. Estimates with an RSE of 50% or more are not published by DAWN. (See also Precision and Standard error.)

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 in order 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 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 American Hospital Association (AHA) Annual Survey of Hospitals.

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 all drug-related ED visits at the responding hospitals selected for the sample.

Sampling weights:
Numeric coefficients used to derive population estimates from a sample.

Seeking detox:
This category of drug-related ED visits captures patient seeking substance abuse treatment, drug rehabilitation, or medical clearance for admission to a drug treatment or detoxification unit. They are classified separately because they often reflect administrative practices that vary across hospitals and may vary over time within the same hospital. Seeking detox visits tend to be concentrated in those facilities that operate specialized inpatient units providing substance abuse treatment or detoxification services, and the largest numbers are found in facilities that require medical clearance for entry into such treatment to be granted in their EDs.

Single-drug case:
A single-drug case is one in which only one drug was involved. Because multiple substances may be recorded for each DAWN case (see Drug), readers should be cautious in interpreting the relationship between a given drug and the number of associated visits or deaths. For example, if the source record for a patient/decedent documented marijuana use, this does not mean that marijuana was the only drug involved in the visit/death or that the marijuana caused the visit/death. One should always consider whether and how many other drugs were used in combination. Even then, attributing a causal relationship between the visit/death and a particular drug may not be possible. DAWN only captures single-drug visits/deaths involving alcohol if the decedent was younger than age 21.

Standard error (SE):
A measure of the sampling variability or precision of an estimate. The SE of an estimate is expressed in the same units as the estimate itself. For example, an estimate of 10,000 visits with an SE of 500 indicates that the SE is 500 visits.

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 (A) two types of hospital ownership (public, private) and (B) up to four size categories (measured in terms of annual ED visits), creating 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, then by state, 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 category of drug-related ED visits captures suicide attempts (e.g., "attempted suicide," "tried to kill self") documented in the medical record in which drug use was involved, including non-medical use of prescription or OTC pharmaceuticals. Suicidal gestures, thoughts, or ideation, including attempts to "harm" self, are assigned to another case type.

Type of case:
A classification used to group similar DAWN cases from the diverse set of all drug-related ED visits. Each case is coded into one and only one category, the first that applies from the following hierarchy: suicide attempt, seeking detox, alcohol only (age < 21), adverse reaction, overmedication, malicious poisoning, accidental ingestion, and other. The rules for assignment of DAWN cases to types of cases are defined in the DAWN ED Decision Tree.

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 emergency departments 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 American Hospital Association's Annual Survey Database.

APPENDIX D
POPULATION DATA

 

Table D1
Population by age and gender: 20041
Age Total U.S. Males Females
Total 293,655,404 144,537,408 149,117,996
   0-5 years 23,923,026 12,232,272 11,690,754
   6-11 years 23,985,999 12,277,377 11,708,622
   12-17 years 25,368,973 12,995,499 12,373,474
   18-20 years 12,350,179 6,358,864 5,991,315
   21-24 years 16,894,923 8,697,646 8,197,277
   25-29 years 19,560,906 9,994,814 9,566,092
   30-34 years 20,471,032 10,341,219 10,129,813
   35-44 years 44,108,652 22,033,881 22,074,771
   45-54 years 41,618,805 20,452,673 21,166,132
   55-65 years 29,078,924 13,999,433 15,079,491
   65 years and older 36,293,985 15,153,730 21,140,255
1 Population estimates for 2004, as of July 2005, from U.S. Census Bureau County Population Dataset CO-EST2004-ALLDATA (see http://www.census.gov/popest/counties/files/CO-EST2004-ALLDATA.csv).
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

APPENDIX E
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.17 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. For data collections, such as DAWN, where self-identification of the individual is not feasible, 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 complete DAWN ED case form, which adopted the combined format in 2003, is reproduced in Figure 2 in this report.

Despite the increased detail allowed by the new categories and the provision for multiple entries, the actual race/ethnicity data extracted from source records and submitted to DAWN is quite limited. This is because the source documents (that is, the ED medical records from which DAWN data are abstracted) rarely contain such detailed information on race/ethnicity of patients.

For reference, estimates of drug-related ED visits by race/ethnicity are presented in Table E1. This analysis, which is based on the most detailed coding of race/ethnicity in DAWN case reports, reveals that estimates for the following categories are too small to be meaningful:

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.18 All cases reported to DAWN as Hispanic or Latino ethnicity are tabulated as Hispanic race/ethnicity, regardless of race.

This lack of detailed race and ethnicity data in DAWN case reports also prevents us from generating rates per 100,000 population for race and ethnicity categories. Data from the 2000 decennial Census were collected and are being tabulated according to the revised race and ethnicity protocol and are therefore incompatible with DAWN estimates.

 

Table E1
Drug-related ED visits, by detailed race/ethnicity: 2004
Race/ethnicity Estimated visits1,2,3
Total drug-related ED visits 1,997,993
One race/ethnicity 1,989,924
   White 1,148,616
   Black/African American 366,017
   Hispanic 167,679
   Asian 4,740
   American Indian/Alaska Native 9,578
   Native Hawaiian/Other Pacific Islander 2,373
   Race Unknown 290,921
Two races/ethnicities 8,060
   White + Black/African American 821
   White + Hispanic 6,506
   White + Asian 40
   White + American Indian/Alaska Native 102
   White + Native Hawaiian/Other Pacific Islander
   Black/African American + Hispanic 359
   Black/African American + Asian
   Black + Native Hawaiian/Other Pacific Islander
   Black/African American + American Indian/Alaska Native
   Hispanic + Asian
   Hispanic + American Indian/Alaska Native 30
   Asian + American Indian/Alaska Native
   Asian + Native Hawaiian/Other Pacific Islander
   American Indian/Alaska Native + Native Hawaiian/
   Other Pacific Islander
166
Three races/ethnicities
   White + Black/African American + Asian
   White + Latino + American Indian/Alaskan Native
Six races/ethnicities
   White + Black + Latino + Asian + American Indian/Alaska Native +
   Native Hawaiian/Other Pacific Islander
1 There are estimates of ED visits based on a representative sample of non-Federal, short-stay hospitals with 24-hour EDs in the U.S.
2 Estimates are all expressed in visits.
3 Three dots (…) indicate that an estimate with an RSE greater than 50% has been suppressed or an estimate less than 30 has been suppressed.
SOURCE: Office of Applied Studies, SAMHSA, Drug Abuse Warning Network, 2004 (09/2005 update).

 


End Notes

1 For more information about the redesign of DAWN, see Substance Abuse and Mental Health Services Administration, Office of Applied Studies. 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, 2004. This and other DAWN publications are available online at http://DAWNinfo.samhsa.gov/.

2 The confidence interval accounts for the margin of error of the estimate. It indicates, with a high degree of confidence, that the true population value was between 1,708,205 and 2,287,781 drug-related ED visits.

3 This review is conducted by data collectors called "DAWN reporters."

4 That is, patients with a history of drug use (and no recent use) are excluded.

5 The classification of drugs used in DAWN is derived from the Multum Lexicon, Copyright 2004, Multum Information Services, Inc. The classification has been modified to meet DAWN's unique requirements (2004). 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.

6 Population estimates for 2004, as of July 2005, from U.S. Census Bureau County Population Dataset CO-EST2004-ALLDATA (see http://www.census.gov/popest/counties/files/CO-EST2004-ALLDATA.csv).

7 In 2004, only reports of amphetamines, cathinone, dimethoxymethamphetamine, and methcathinone are classified in this category. Drugs specifically identified as amphetamine-dextroamphetamine, benzphetamine, or dextroamphetamine are now classified as CNS stimulants. This is a change from 2003 when all these drugs were classified as stimulants.

8 This includes only single-ingredient formulations. Many multi-ingredient pharmaceuticals containing diphenhydramine are classified elsewhere, e.g., as respiratory agents.

9 This is not wholly unexpected since the numbers of seeking detox ED visits can vary dramatically across hospitals, and the presence of specialized detoxification or substance abuse treatment units is not accounted for in the DAWN sample design.

10 AHA Annual Survey Database, Fiscal Year 2001 Health Forum LLC, Copyright 2003, One North Franklin Street, Chicago, IL 60606.

11 Metropolitan Statistical Area is one category of Core Based Statistical Area (CBSA). The other CBSA category is the Micropolitan Statistical Area.

12 Substance Abuse and Mental Health Services Administration, Office of Applied Studies. 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, 2004.

13 This design took into account that expansion into additional metropolitan areas would occur over a period of time, but it has been similarly useful for contraction.

14 Four Census regions times two ownership categories times three size categories equals 24 strata.

15Substance Abuse and Mental Health Services Administration, Office of Applied Studies. Drug Abuse Warning Network, 2003: Interim National Estimates of Drug-Related Emergency Department Visits, Appendix B, DAWN Series D-26, DHHS Publication No. (SMA) 04-3972, Rockville, MD, 2004.

16 Population estimates for 2004, as of July 2005, from U.S. Census Bureau County Population Dataset CO-EST2004-ALLDATA (see http://www.census.gov/popest/counties/files/CO-EST2004-ALLDATA.csv).

17 See Office of Management and Budget, Revisions to the Standards for the Classification of Federal Data on Race and Ethnicity, Federal Register, 62 FR 58782, October 30, 1997.

18 One exception is that if two races are reported and the second is reported as unknown, the episode is coded for the known race.

Click here to go to the top of this pageClick here to go to the Table of Contents for the State Estimates from the 2003-2004 NSDUH