|
This glossary defines terms used by the Drug Abuse Warning Network
(DAWN) in data collection activities, analyses and publications.
DAWN collects data and publishes findings separately for emergency
departments (EDs) and death investigation jurisdictions. As a result,
there are a number of terms that are unique to each component of
DAWN.
Terms with this symbol are applicable to the emergency department
component.
Terms with this symbol are applicable to the medical examiner/coroner
component.
Click on a letter to move to that section of the Glossary, or scroll
through all entries below.
A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P |
Q | R | S | T | U | V | W | X | Y | Z |
Cause
of death: Cases are reportable to DAWN if the death investigation
concludes that the death was either directly or indirectly caused
by drug abuse. If a death was directly caused by drug abuse (e.g.,
a drug overdose), DAWN refers to the death as drug-induced.
If drug abuse was a contributing factor in the death, but not the
immediate or sole cause, then DAWN refers to the death as drug-related.
It is important to note that DAWN data include both types of deaths.
It is also important to note that a drug-induced death may involve
more than a single drug. (See Single-drug episode.)
Certified death: Any case accepted and
reviewed by a medical examiner or coroner, who uses information
from the death investigation to complete the death certificate.
Consistent panel: DAWN does not impute
missing data for ME/C jurisdictions that have not reported for all
or part of a given year. Therefore, tables and charts showing trends
in deaths over time are based on a "consistent panel"
of reporting jurisdictions. A consistent panel includes those jurisdictions
that have reported data for at least 10 months of each year reflected
in the trend table/chart. The reason for a consistent panel is to
ensure that apparent changes over time are not a result of gaps
in reporting. Because participating jurisdictions may change from
year to year, consistent panels used in published reports will also
change from year to year. This means that trends published in one
annual report are not necessarily comparable to trends published
in subsequent annual reports.
Coroner: Death investigation jurisdictions
typically use either a medical examiner system or a coroner system.
Unlike medical examiners, coroners need not be physicians; usually
the only prerequisite for serving as a coroner is that the individual
be more than 18 years of age and a resident of the county or district
to be served. Coroners are typically elected rather than appointed.
They may have jurisdiction over counties or districts within states.
(See also Jurisdiction and Medical examiner.)
Coterminous U.S.: The contiguous 48 continental
States and Washington, DC. Excludes Alaska and Hawaii. National
estimates from DAWN refer only to the coterminous U.S.
Disposition of ED patient: Suggestions
or recommendations made or actions taken by the hospital as they
relate to the patient's presenting problem:
- Treated and released or referred The patient was
given appropriate ED treatment and was released or, after appropriate
ED treatment, the hospital referred the patient to another agency
or to a private physician for additional services.
- Admitted to hospital The patient was admitted
as an inpatient to a hospital.
- Left against medical advice The patient left the
treatment setting without a physician's approval.
- Died The patient expired.
Drug abuse: The nonmedical use
of a substance for any of the following reasons: psychic effect,
dependence, or suicide attempt/gesture. In DAWN, nonmedical use
means:
- the use of prescription drugs in a manner inconsistent with accepted
medical practice;
- the use of over-the-counter drugs contrary to approved labeling;
or
- the use of any substance (e.g., heroin, marijuana, peyote, glue,
aerosols) for psychic effect, dependence, or suicide.
Drug abuse episode: A reported ED visit
that involved drug abuse. Episodes involving patients under the
age of 6 or over the age of 97 are not reported to the DAWN system.
The number of ED patients in DAWN is not synonymous with the number
of patients involved. One patient may make repeated visits to an
ED or to several EDs, thus producing a number of episodes. It is
impossible to determine the number of unique patients involved in
the reported ED episodes because no patient identifiers are collected.
Drug category: A generic grouping
of substances reported to DAWN, based on the classification of generic
drugs by 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.
The DAWN system has accumulated a vocabulary of thousands of substance
names that have been mentioned in incidents of abuse. This vocabulary
is updated monthly by the inclusion of new abuse substances and,
through receipt of identifying information, the reclassification
of drugs. Occasionally, this reclassification may result in a drug
being shifted to a different drug grouping. The DAWN drug groupings
are periodically reviewed in order to reflect the most recent changes
in pharmaceutical classifications and drug legislation. Occasional
changes in drug classification should be taken into consideration
when comparing drug data from this publication with other DAWN publications.
These classifications may involve street names and brand names,
which are sometimes used to identify a substance and its generic
drug group. Individual drugs comprising the most commonly reported
drug categories can be found in Tables 2.3 to 2.7 of Emergency
Department Trends from DAWN.
Additional clarification is provided for the following drug categories:
- Alcohol-in-combination DAWN does not gather data
on alcohol used alone, only alcohol used concomitantly with another
abused substance. Therefore, all alcohol mentions are combination
mentions.
- Club drugs During the 1990s, use of certain illicit
drugs was linked to "raves" and dance clubs. These substances
are commonly referred to as "club drugs." When used
in DAWN, the term "club drugs" includes Ketamine, flunitrazepam
(Rohypnol), gamma-hydroxy butyrate (GHB, or its precursor, gamma
butyrolactone [GBL]), and methylenedioxymethamphetamine (MDMA
or Ecstasy). Although commonly used in the rave scene, methamphetamine
and hallucinogens are classified separately from club drugs in
DAWN.
- Drug unknown "Drug unknown" may be recorded
when drug abuse was known or suspected to have been involved,
but the specific substance could not be determined.
- Heroin and Heroin/morphine This is the only drug
classified differently in the ED and mortality components of DAWN.
In the ED publications, heroin is classified as a major substance
of abuse, separate from morphine, which is classified as a narcotic
analgesic under central nervous system (CNS) agents. In the mortality
data publications, heroin and morphine are classified together
in a single category. When heroin is ingested, it is metabolized
to morphine, so that the toxicology testing commonly used in death
investigations often does not distinguish between the two. Therefore,
a mention of either substance is recorded as heroin/morphine.
A case mentioning both heroin and morphine will be "de-duplicated"
and counted as a single heroin/morphine mention.
- Inhalants This category includes anesthetic gases
and psychoactive nonpharmaceutical substances for which the documented
route of administration was inhaled, sniffed, or snorted. Psychoactive
nonpharmaceuticals fall into one of the following 3 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.
- Major Substances of Abuse We use this term to
refer to the most commonly abused drugs (e.g., alcohol-in-combination
and cocaine) and those drugs that are typically referred to as
"illicit."
- Other Substances of Abuse We use this term to
refer to pharmaceutical agents not included in the Major Substances
of Abuse.
Drug combinations: Published tables from
the DAWN mortality data refer to "drug combinations" rather
than "drug concomitance" (the term used in the
ED component). This term refers to multiple drug mentions for a
single death, and tables show particular combinations of substances
reported for deaths. Readers should note that DAWN cannot differentiate
between drugs actually used in combination (simultaneously) and
drugs used sequentially.
Drug concomitance: This term refers to
whether a drug abuse episode involved a single drug (one mention)
or multiple drugs (multiple mentions).
Drug-induced death: A death directly resulting
from drug abuse or other substance abuse, such as drug overdoses
or the interactive effects of drug combinations. When more than
one drug is mentioned, it cannot be determined which or whether
one drug was the sole and direct cause of the death.
Drug mention - This refers to
a substance that was recorded ("mentioned") in a DAWN
case report. In addition to alcohol-in-combination, up to 4 substances
("mentions") can be reported for each ED episode, and
up to 6 substances can be reported for each drug abuse death. Therefore,
the total number of drug mentions exceeds the total number of ED
visits or deaths. Even when only one drug is mentioned, it should
not be assumed that the substance was the sole and direct cause
of the episode or death; allowances should be made for reportable
drugs not mentioned or other contributory factors. (See also Single-drug
episode/death.)
Drug-related death: A death in which the
abuse of a drug is a contributing factor, but is not the sole cause
of death. Such cases include drug abuse that exacerbates a pre-existing
physiological condition; drug abuse in combination with an
external physical event (e.g., a fall or automobile accident);
or a medical disorder that was itself caused by drug abuse
(e.g., hepatitis contracted through injection drug use). Drug-related
deaths are classified into two types, confirmed and presumed.
The drug-relatedness is "confirmed" if documentation in
the decedent's file substantiates that conclusion. The drug-relatedness
is "presumed" if the investigation suggests drug involvement,
but the medical examiner/coroner has insufficient evidence to list
drug abuse as a contributing cause on the death certificate. Both
confirmed and presumed deaths are included in the published mortality
data tables.
Drug use motive: DAWN classifies ED drug
abuse episodes according to one or more of the following reasons
for taking a substance(s):
- Psychic effects A conscious action to use drugs
to improve or enhance any physical, emotional, or social situation
or condition. Two categories of psychic effect are:
- Use of drugs for experimentation or to enhance a social situation
(e.g., curiosity, peer pressure, "just wanted to know what
it felt like," "wanted to have fun," "to
get high," "for kicks," "to party");
and
- Use of drugs to improve or enhance any mental, emotional,
or physical state (e.g., depression, anxiety, to relieve headache,
reduce pain, stay awake, lose weight, relax, help study, get
to sleep). Referred to in DAWN as "other psychic effects."
- Dependence A physiological or psychological condition
characterized by a compulsion to take the drug on a continuous
or periodic basis in order to experience its effects or to avoid
the discomfort of its absence (e.g., had to take, had to have,
needed a fix).
- Suicide attempt or gesture Successful or unsuccessful
action(s) taken for the purpose of self destruction or to gain
attention.
- Other reason Used when the reason for taking the substance
cannot be classified into the categories above.
Estimate: A statistical estimate is the
value of a parameter (such as the number of drug-related ED episodes)
for the universe that is derived by applying sampling weights to
data from a sample. DAWN produces representative statistical estimates
for 21 metropolitan areas based on data from a sample of EDs in
each of the 21 areas. An estimate for the coterminous U.S. is produced
by summing estimates for the 21 metropolitan areas and an estimate
for the National Panel.
Form in which drug was acquired: The form
in which the substance was received by the user/abuser, not the
form in which the substance was consumed. Categories are: tablet/capsule/pill,
aerosol, liquid, powder/crystal, paper, pieces/chunks, injectable
liquid, cigarette, plant material, unknown, and other. Note: this
information is often not documented in ED records and is therefore
missing in DAWN tabluations. Caution should therefore be exercised
in interpreting this information.
Hospital emergency department (ED): Only
hospitals that meet eligibility criteria for DAWN are recruited
to participate. To be eligible, hospitals must be non-Federal, short-stay,
general medical and surgical facilities with EDs that are open 24
hours a day, 7 days a week, and located in the coterminous U.S.
Specialty hospitals; hospital units of institutions; long-term care
facilities; pediatric hospitals; hospitals operating part-time emergency
departments; hospitals in Alaska and Hawaii; and hospitals operated
by the Veterans Health Administration and the Indian Health Service
are excluded.
Jurisdiction: DAWN uses the term "jurisdiction"
to mean the geographic area for which a medical examiner/coroner's
office is responsible. In many states, there is a 1:1 correspondence
between jurisdictions and counties. In some states, there are multiple
medical examiner/coroner offices within a given county, or there
may be multiple counties covered by a "district" that
includes one or more medical examiners/coroners. A few states are
organized as a single statewide jurisdiction.
Understanding jurisdictions is important because this assists readers
in interpreting aggregated data. Published DAWN mortality data are
aggregated into metropolitan areas, which often comprise multiple
jurisdictions. In some states, there are different death investigation
procedures for different jurisdictions (most notably, some jurisdictions
have medical examiner systems, while others have coroner systems).
There are nearly always some differences in death investigation
procedures across states (and notably, some metropolitan areas include
jurisdictions in multiple states). Readers should be mindful of
these variations when interpreting or comparing data.
Information on death investigation practices and an updated list
of jurisdictions throughout the U.S. and Canada are available from
the Centers
for Disease Control's Epidemiological Program Office.
Manner of death: This variable is used
to describe how the decedent died. It is applicable to both drug-induced
and drug-related deaths. On the DAWN data collection form, manner
of death is coded into the following categories:
- Accidental/Unexpected Although the drug abuse was deliberate,
the resulting death was unintended.
- Suicide Death in which there is evidence that
the decedent deliberately used drugs to bring about his/her demise.
- Homicide Death in which the decedent's life was
taken by another individual by means of drugs. These cases, which
do not involve the intentional abuse of drugs by the decedent,
are not currently included in published tabulations of
DAWN mortality data.
- Natural Death was due to natural causes such as
a medical disorder or disease process, if drug abuse caused or
worsened the decedent's condition.
- Undetermined The manner of death cannot be determined
from all available evidence.
In Mortality Data from DAWN, manner of death is collapsed
into three categories: suicide, accidental/unexpected, and "All
others." The "All others" category includes cases for
which manner of death was recorded as natural, unknown, or undetermined,
and cases for which manner of death was missing.
Medical Examiner (ME): Death investigation
jurisdictions typically use either a medical examiner system or
a coroner system. Most medical examiners are licensed physicians
or forensic pathologists, and are generally appointed (rather than
elected). They may have jurisdiction over a county, district, or
entire state. (See also Coroner and Jurisdiction.)
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. The current DAWN ED sample,
which was redesigned in the 1980s, is based on the definitions of
Metropolitan Statistical Areas (MSAs) and Primary Metropolitan Statistical
Area (PMSAs) issued by the Office of Management and Budget (OMB)
in 1983, with a few exceptions. Metropolitan areas represented in
the DAWN mortality data system are consistent with those represented
in the DAWN emergency department system, also with a few exceptions.
Users of DAWN should note that the emergency department component
provides estimates for each of the 21 metropolitan areas. However,
in the mortality data component, only raw counts are provided, and
in many instances less than 100% of the MSA is represented in those
counts.
National Panel: This term is used to denote
2 concepts relative to DAWN ED data: (1) The universe of eligible
hospitals outside the 21 DAWN metropolitan areas but within the
coterminous U.S. and (2) The sample of hospitals in DAWN that were
selected from this universe. The National Panel sample is weighted
to produce estimates for the National Panel universe. (See also
Metropolitan area.)
Not otherwise specified (NOS):
Catch-all category for substances that are not specifically named
in the listing. 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 and death
investigation case files).
Not tabulated above (NTA): Designation
used when categories are not presented in complete detail; smaller
units are combined in the NTA category.
p-value: The probability value is the actual
probability associated with a statistical estimate; this is then
compared with the significance level to determine whether that value
is statistically significant. For a statistically significant result,
the p-value must be less than or equal to the significance
level. The traditional significance levels are p less than 0.001,
0.01, 0.05, and 0.10. A result with a p-value less than 0.05 is
considered statistically significant in DAWN ED publications.
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
an RSE of 50 percent or higher 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: Beginning in
January 2000, the race and ethnicity categories collected on DAWN
case report forms changed to match a change in the standard protocol
issued by the Office of Management and Budget in 1997. (Revisions
to the Standards for the Classification of Federal Data on Race
and Ethnicity, Federal Register, 62 FR 58782, October 30, 1997.)
The new protocol permits 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 2 categories,
"Asian" and "Native Hawaiian or Other Pacific Islander");
and elimination of the "Other" category.
The race/ethnicity categories on the DAWN data collection forms
are as follows:
- Race
- 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.
- 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.
- 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.
- Native Hawaiian or Other Pacific Islander A
person having origins in any of the original peoples of Hawaii,
Guam, Samoa, or other Pacific Islands.
- Unknown Used when documentation of race is not
available from source records.
- Ethnicity
- Hispanic or Latino A person of Cuban, Mexican,
Puerto Rican, South or Central American, or other Spanish culture
or origin, regardless of race.
- Not Hispanic or Latino Ethnicity does not meet
the definition of Hispanic or Latino.
- Unknown Used when documentation of ethnicity
is not available from source records.
Tabulations of Race/Ethnicity Data. Despite the increased
detail allowed by the new categories, the actual race/ethnicity
data reported to DAWN changed very little because race and ethnicity
are often not documented with this level of specificity in patient/decedent
records. As a result, we have retained the classification used previously
to tabulate DAWN data. The one exception is that we now collapse
the less commonly used categories into a category termed "Not
tabulated above (NTA)" instead of "Other." Categories
used to tabulate race and ethnicity data in the ED publications
are:
- 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 is placed in the category Hispanic, regardless of race.
- Race/ethnicity not tabulated above (NTA) This
includes those categories that are too small to report independently
including: two or more races, American Indian or Alaska Native,
Asian, Native Hawaiian or Other Pacific Islander.
- Unknown Race and ethnicity are unknown. Those
who are identified only as Hispanic are classified as Hispanic.
In Mortality Data from DAWN, race/ethnicity data are tabulated
as White, Black, Hispanic, and All others, where "All others"
includes other reported races and ethnicities as well as unknown
or missing data.
Rank: A rank indicates the relative
frequency of a measure, such as mentions for a particular drug category.
For example, a drug category ranked second indicates that it accounted
for the second highest number of mentions among all drug categories.
When 2 or more drugs receive equal numbers of mentions, they are
assigned the same rank. A difference in rank should be considered
only as indicative of a difference in frequency among drugs reported
to DAWN, regardless of the size of the difference. Such differences
are not necessarily meaningful or statistically significant.
Reason for present ED contact: The reason
for the patient's visit to the ED, based on documentation provided
in the medical record. Categories are:
- Overdose/toxic ingestion Either intentional or
accidental (e.g., effects of suicide attempt, coma). Anyone whose
reason for contact is overdose is placed in this category, regardless
of other reasons.
- Unexpected reaction The drug's effect was different
than anticipated, thus causing concern (e.g., bad trip, panic,
hallucinations).
- Withdrawal Symptoms which occur when a patient
stops taking a substance upon which she/he is physiologically
dependent and suffers physical symptoms, including abdominal pain,
cold sweat, hyperactivity, and tremors that require treatment.
- Chronic effects Secondary conditions resulting
from habitual use or dependence, including malnutrition, tetanus,
blood poisoning, and so forth.
- Seeking detoxification Patients with identified
problems with chronic substance abuse who seek admission to a
detoxification program and receive treatment from emergency department
staff. This category was added to the data collection form in
1987. Some hospitals require patients to be processed in the ED
prior to admission for detoxification. Caution should therefore
be exercised in interpretation of this category and the remaining
information.
- Accident/injury Injuries resulting from accidents
that were caused by or related to drug abuse. This category was
added to the data collection form in 1987.
- Other Reasons which cannot be classified into
one of the aforementioned categories.
Reason for taking substance: See Drug
use motive.
Relative standard error (RSE): A measure
of the sampling variability or precision of an estimate defined
as the estimate's standard error (SE) expressed as a percentage
of the estimate's value. For example, an estimate of 2,000 cocaine
mentions with an SE of 200 mentions has an RSE of 10 percent. (See
also Precision and Standard error.)
Route of drug administration:
DAWN reporters are asked to record the method by which the substance
was taken into the drug abuser's body according to the following
categories:
- Oral Substance was ingested through the mouth
(swallowed).
- Injection Substance entered the body through a vein (intravenously),
into the muscle (intramuscularly), or under the skin (subcutaneously).
- Inhaled Gases or fumes of a substance were taken
into the body by inhaling through the nose or mouth into the lungs
(e.g., inhaling the fumes of glue, aerosols, paints, gasoline).
- Smoked (includes freebase) Substance was consumed
by smoking a cigarette, pipe, or similar device.
- Sniffed/snorted Substance, acquired in a powder
or crystalline form, was forcefully inhaled through the nose.
- Other This category is used when the route of
administration of the substance cannot logically be included as
any of the above.
Note: this information is often not documented in patient/decedent
files and is therefore missing in DAWN tabulations. Caution should
therefore be exercised in interpreting this information.
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 episodes at the responding
hospitals selected for the sample.
Sampling weights: Numeric coefficients
used to derive population estimates from a sample.
Single-drug episode/death: A
single-drug episode or death is that in which only one drug was
involved. Because multiple substances may be recorded for each DAWN
case (see Drug mention), readers should exercise caution
in interpreting the relationship between a given drug and the number
of associated ED visits or deaths. For example, if records for a
given patient "mentioned" marijuana, this does not mean
that marijuana was the only drug involved in the ED visit or that
the marijuana caused the ED visit. One should always consider whether
and how many other drugs were used in combination, but even then
attributing a causal relationship between the visit and a particular
drug may not be possible. Additionally, because alcohol is only
documented if used in combination with another drug, DAWN cannot
provide single-drug episode/death totals for alcohol.
Source of substance: The immediate source
of the substance that the patient abused is coded as follows:
- Patient's own legal prescription This is coded
only when the abuser was legally prescribed the drug of abuse.
If one patient obtains a drug by legal prescription and sells
it to another who abuses it, the source to the abuser is marked
"street buy." If the patient for whom the prescription
was issued gives the drug to another patient who abuses it, the
source to the abuse is "other unauthorized procurement."
- Street buy The drug abuser purchased a drug and/or
prescription from a source other than legitimate channels.
- Other unauthorized procurement The drug was acquired
in a manner not consistent with accepted medical care but was
not bought on the street. This category includes drugs purchased
using forged prescriptions, stolen, or received as a gift.
- Other Used when the source of the substance cannot
logically be included as any of the above. This category includes
all over-the-counter medications.
- Unknown - Reported when information on source was unavailable.
Note: this information is often not documented in ED records and
is therefore missing in DAWN tabulations. Caution should therefore
be exercised in interpreting this information.
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 cocaine mentions with an SE of 500 indicates
that the SE is 500 mentions.
Strata (plural), stratum (singular): Subgroups
of a population 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 21 metropolitan area cells plus an additional cell for the
National Panel. Then, within these cells strata are defined according
to the annual number of ED visits, whether the hospital is located
inside or outside the central city of the metropolitan area, and
by the presence or absence of an organized outpatient department,
alcohol/chemical dependence inpatient unit, or both. The strata
are as follows:
Stratum |
Annual ED visits |
Location within metropolitan area |
Outpatient department or alcohol/chemical
dependence inpatient unit |
In the 21 DAWN metropolitan areas: |
0 |
>80,000 |
Not applicable |
Not applicable |
1 |
<80,000 |
Central city |
Both |
2 |
<80,000 |
Central city |
One only |
3 |
<80,000 |
Central city |
Neither |
4 |
<80,000 |
Outside Central city |
Both |
5 |
<80,000 |
Outside Central city |
One only |
6 |
<80,000 |
Outside Central city |
Neither |
In the National Panel: |
0 |
>80,000 |
Not applicable |
Not applicable |
7 |
<80,000 |
Not applicable |
Both |
8 |
<80,000 |
Not applicable |
One only |
9 |
<80,000 |
Not applicable |
Neither |
Note: Stratum "0" is defined for each of the 21 metropolitan areas and the National Panel cells. See Drug Abuse Warning Network Sample Design and Estimation Procedures: Technical Report, November 1997. |
Statistically significant: A difference
between 2 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.)
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 coterminous U.S. with EDs open 24 hours a day, 7 days a week.
(See also Coterminous U.S.).
|