Click to see the modal popup.

Technical Assistance and Training - Data Sources

SAMHSA Grants - Data Sources to Guide Your Grant Application

Data are essential for planning and evaluating mental health and substance abuse treatment and prevention programs. However, in many instances relevant data are not available. The problem is particularly severe with respect to geographical areas smaller than States where little information on substance abuse problems exists outside the criminal justice system and the occasional school survey.

The Substance Abuse and Mental Health Services Administration (SAMHSA) has developed several data sets that might be useful under certain conditions to assess substance abuse and mental health problems and trends at the State level and in some cases even smaller geopolitical areas.  A review of these data sets may also be instructive with respect to the kinds of data that can be used to inform policy and programs and methods for collecting such information.  (

Drug Abuse Warning Network (DAWN)The Drug Abuse Warning Network (DAWN) is a national surveillance system that monitors emergency department (ED) visits associated with drug abuse as well as drug-related deaths that are investigated by medical examiners and coroners.  DAWN data are used by Federal and local agencies, treatment providers, and epidemiologists to track trends in existing and emerging drugs of abuse.

From the 1988 to 2002, DAWN relied on a sample of about 550 hospital-based emergency departments to provide data on ED visits related to substance abuse.  Non-Federal, short-stay, general hospitals that operate 24-hour/7-day-a-week EDs were eligible for the sample.  This sample was designed to produce estimates of the number of drug abuse-related ED visits in the coterminous U.S. (48 States and the District of Columbia) and in 21 major metropolitan areas.  Mortality data were collected from a non-random set of medical examiners and coroners.  Drug-related deaths could be reported at the metropolitan and jurisdiction (county) levels, but no national estimates were possible.

Because DAWN data are collected continuously and contain extensive drug detail, they can be used to monitor trends, warn of the emergence of dangerous new drug problems (such the increase in use of ecstasy and methamphetamine), identify the abuse potential of prescription and over-the-counter drugs, and describe changing patterns of drug abuse in local communities.  DAWN is the only national data collection system that has the capacity to identify specific and infrequently used substances as they emerge and diffuse across segments of the population and geographic areas.  DAWN also provides a model for establishing local surveillance systems to monitor substance abuse trends and patterns.

Beginning in 2003, a myriad of changes have been introduced to DAWN with the aim of improving the consistency and utility of the data.  Another consequence of these changes is a permanent disruption in trends.  The types of cases reportable to DAWN have been expanded to include all types of drug-related ED visits and deaths, not just those with documented drug abuse.  With the expansion of cases to include use, misuse, and abuse of drugs, new data items were added to better characterize the health conditions and causes of death associated with drug use.  The DAWN sample of hospitals also has been redesigned, and a multi-year expansion into additional metropolitan areas will improve geographic and population coverage.  When the expansion is complete, DAWN will be capable of producing estimates for the entire Nation as well as estimates for 48 metropolitan areas.  Expansion of the medical examiner and coroner component of DAWN will mirror the metropolitan-area expansion for hospitals.  No national estimates for drug-related deaths will be possible; instead, the objective is to achieve a census of jurisdictions within targeted metropolitan areas.

Each facility that participates in DAWN has a designated DAWN reporter who reviews source records to identify substance abuse cases.  A case is reported to DAWN if there is evidence that an illicit drug, prescription or over-the-counter medication, dietary supplement, or non-pharmaceutical inhalant was used in a suicide attempt, due to dependence, or to achieve psychic effects.  Patient demographic and episode characteristics, including the specific drugs involved, at abstracted from the source records and submitted to DAWN.

Public use data are not available from DAWN due to confidentiality restrictions.  Published reports from DAWN are available online at

Drug And Alcohol Services Information System (DASIS)

The Drug and Alcohol Information System (DASIS) is the only comprehensive national source of information on the services available for substance abuse treatment and the number and characteristics of individuals admitted to treatment.

DASIS consists of three data sets. The Inventory of Substance Abuse Treatment Services (I-SATS) is a list of all organized substance abuse treatment programs in the United States that SAMHSA has been able to identify. The list is assembled from a variety of sources including the Yellow Pages and business directories. In calendar year 2003 the list contained the names and addresses of approximately 18,000 facilities. The I-SATS is used as a list frame for the National Survey of Substance Abuse Treatment Services (N-SSATS).

The N-SSATS collects information on all treatment facilities in the inventory and is conducted annually.  N-SSATS obtains information on facility ownership, treatment services, and number of patients in treatment on a given day.  The survey is conducted by mail and via the Internet.  Facilities that do not complete a paper questionnaire or web questionnaire are contacted by telephone to complete the questionnaire in computer assisted telephone interviews.  In 2002, 57 percent of treatment facilities completed the mail questionnaire, 20 percent completed by web and an additional 23 percent participated in the telephone survey.

The State substance abuse agencies identify those facilities in the survey that are certified, licensed, or otherwise recognized as legitimate sources of treatment (approximately 80% of the facilities responding to the survey).  The names, addresses, and services offered by these facilities are included in a National Directory of Drug and Alcohol Abuse Treatment Programs which is published each year.  This information also appears in a Web-based version of the Directory known as the Substance Abuse Treatment Facility Locator.  The Locator allows individuals and families seeking treatment to find the most convenient and appropriate services for their needs ( ).

In contrast to the printed Directory, the computer-based version is updated monthly. This is important because more than 20% of treatment facilities report some change in telephone number, address or services during the year. Because of the relatively comprehensive nature of the Web-based facility locator system, N-SSATS provides a dependable source of information on treatment services in all areas of the country. The system also can be used as a basis for other data collection efforts at the local level.

The third component of DASIS is the Treatment Episode Data Set (TEDS). TEDS contains information on substance abuse treatment admissions that is routinely collected by States in monitoring substance abuse their treatment programs. Selected data items from individual State data files are converted to a standard format consistent across States to create a common data base. In 2000, the most recent year for which complete data are available, there were approximately 1.7 million admissions reported to TEDS.  TEDS includes admissions primarily from facilities that receive some sort of public funding through state substance abuse agencies.  Data on admissions to Federally owned/operated facilities are not reported to TEDS.

TEDS consists of a Minimum Data Set collected by all States and a Supplemental Data Set collected by some States. Data items for each admission include demographic information, substances of abuse and route of administration, information on prior treatment episodes, and the treatment plan. TEDS now includes a discharge data set as well. As of 2003, almost half of the states report discharge data to TEDS.  TEDS can be used to describe trends in treatment admissions to publicly-funded facilities at various geographic levels. 

The DASIS data sets provide the basis for identifying treatment programs, describe services both locally and nationally, and enumerate and characterize persons admitted for treatment. Because N-SSATS and its predecessors have been in place for almost 20 years, it is possible to track changes over time in the structure and composition of the treatment industry and the utilization of services. This type of information provides a basis at both the national and local levels to assess trends in drug use, treatment, and the impact of national and local policies on substance abuse.

DASIS reports and information can be accessed by going to the SAMHSA web site ( and clicking on statistics/data and then the button for DASIS. Access to DASIS data sets can be obtained by clicking on statistics/data and the button for SAMHDA (Substance Abuse and Mental Health Data Archive).  The database archive can also be reached by going to the web page (

National Mental Health Reporting Program

Recurrent Surveys:

The Biennial Survey of Mental Health Organizations (SMHO) B A biennial enumeration survey of specialty mental health organizations, with a more detailed sample survey, that collects organizational-level characteristics (e.g., ownership, range of services provided, caseload data, staffing, revenues and expenditures).  Data from the SMHO  have proven to be critical in cost estimation for mental health care planning, especially for community-based services.  The 1998 survey includes, for the first time, a sample survey of managed behavioral healthcare organizations.  Data collected from these organizations include characteristics of behavioral healthcare plans (e.g., number of plans with behavioral healthcare benefits, number of persons with behavioral healthcare benefits, and amount of annual claims paid for behavioral healthcare services). Year 2000 data will be available later in FY2003. 

The Annual Census of Patient Characteristics - State and County Mental Hospital Inpatient Services - Additions and Resident Patients B An annual census that provides national and state level statistics on the age, gender, and race/ethnicity by diagnosis of additions and resident patients in State and county mental hospitals throughout the nation.  The Census is instrumental in monitoring trends in deinstitutionalization.  The number of  forensic units in each State mental hospital is included in the 2000 Census.  Data are available for all years from 1950 through 2000. Data for 2001 will be available later in FY2003.

The Client/Patient Sample Survey B A client-based sample survey which provides national estimates on the sociodemographic, clinical and service use characteristics of persons admitted and persons under care in the inpatient, residential and less than 24-hour care programs of specialty mental health organizations.  The survey updates earlier client/patient sample surveys (the last comprehensive survey was in conducted in 1986), and provides data for trend analyses.  In addition, the 1997 survey design included, for the first time, an oversample of children and youth under age 18.  As a result, the 1997 survey provides national estimates to better address services received by children and youth, and to allow for more in-depth analyses of this population. The next survey is planned for FY2005. 

Single-Time Surveys:

The 1993 Inventory of Local Jail Mental Health Services B An enumeration survey that collects, for the first time, national information on the availability of mental health services within local jails and on the numbers of persons served in these settings.

The 2001-2002 National Survey of Mental Health Mutual Support Groups and Self Help Organizations B The goals of this survey are (1) to provide national estimates of the number of mutual support groups, self-help organizations, and businesses/services run by and for consumers and/or their families; and (2) for these consumer entities, to describe their structure, types of activities undertaken, approaches to well-being and recovery, resources, and links to other community groups.  Data will be available in the Fall of 2003.

National Survey on Drug Use and Health (NSDUH) 

The National Survey on Drug Use and Health (NSDUH), previously known as the National Household Survey on Drug Abuse (NHSDA), is an ongoing survey of the civilian, non-institutionalized population 12 years of age and older. The Survey provides information on the nature and extent of substance use and abuse in the general population including the number and characteristics of persons using alcohol, tobacco, and illicit drugs, and the number of persons in need of or receiving substance abuse treatment. The NSDUH has been the primary source of information on substance abuse problems and treatment since it was initiated in 1971.

The NSDUH collects information by administering questionnaires to members of households, non-institutional group quarters such as shelters, dormitories, and rooming houses, and civilians living on military bases through a face-to-face interview at their place of residence. Beginning with the 1999 NSDUH, the mode of the survey was changed from a system where much of the survey involved respondents completing paper questionnaires to a system where a portable computer is used to both ask and record answers to questions.

Beginning in 1999 the sample employed a 50 State design with an independent, multi-stage area probability sample for each of the 50 states and the District of Columbia. For the eight largest states, which together account for about 48% of the population, the samples are large enough to support direct state estimates. For the remaining 42 states and the District of Columbia, the samples are selected to support state estimates using small area estimation techniques. The total sample is 67,500 which is approximately equally distributed across three age groups: 12-17 years, 18-25 years, and 26 years and older. In 2002 the weighted response rates were 90.7 percent for the household screener and 78.9 percent for the interview.

The NSDUH can provide modeled or direct estimates for the States for some variables.  It is also possible to generate some limited sub-state estimates by combining several years of data.

Public use data sets and published reports of the NSDUH data can be downloaded from the SAMHSA web page ( and then clicking statistics/data. This page will provide the option of obtaining information and reports from the NSDUH or obtaining data sets through the data archive by clicking the button for SAMHSA online database archive. The data archive can also be reached at the web page ( ). 


Last updated: 04/28/2011