64

(Agencies Involved: DOT/NHTSA; HHS/ASPE, SAMHSA, NIH, CDC, FDA; DOJ/DEA, OJP; OMB/OIRA; OSTP) Primary data collected by federal surveys provide a window into the prevalence of drug use and associated behaviors, and lend themselves to extensive analysis to address specific policy questions, such as how many persons use specific drugs, how many need treatment, how similar or different are drug use patterns for racial or ethnic sub-groups, rural or urban populations, parolees and probationers, older adults, and other demographic sub-populations. The National Survey on Drug Use and Health (NSDUH) provides nationally representative data on much of this information on persons living in households.417 The school-based Monitoring the Future study and, to a more limited extent, the Youth Risk Behavior Survey, measure prevalence among youth who are attending school. Other surveys include the new Drug Abuse Warning Network (DAWN), reconstituted in 2018 and anticipated to yield its first full-year, nationally representative data on druginvolved admissions to U.S. emergency departments for calendar year 2020. Prisoner and jail inmate surveys are conducted by the Bureau of Justice Statistics, although less frequently, with the most recent prison inmate survey conducted in 2016.418 Besides primary data from individual respondents, facilities also are surveyed. For example, the National Survey of Substance Abuse Treatment Services (N-SSATS) collects data from substance abuse treatment facilities in the United States on facility location, scope, and characteristics. Many nationally representative household and school surveys lack coverage for populations at high risk of drug use that are outside the realm of their survey universe – such as youth who have quit school, people experiencing homelessness, sex workers, or arrestees. These subgroups are generally smaller, more hidden, and harder to access, and therefore would require more resources and novel approaches to data collection. One limitation of many drug surveys is that they rely on self-report without the resources to obtain additional corroborating information. Because behaviors associated with drug use are illicit, self-reporting can result in under-reporting of use. Ideally self-reporting should be complemented by corroborating data sources, such as measurements relying on biological samples, when these are feasible and can detect use during the period of time targeted by the survey (e.g., past year, past month, past week). Thus far, the prohibitive cost and logistical issues posed by collecting such specimens has limited their routine use. Maintaining existing primary data collection systems is a continuing endeavor in the face of limited or shrinking resources. The discontinuation of the Arrestee Drug Abuse Monitoring (ADAM) program and DAWN, followed by the recent resurrection of the latter are examples of the disruptions that can plague previously established data « « « « « « NATIONAL DRUG CONTROL STRATEGY 109 collection systems. In the long run, it is necessary to retain and improve existing data resources by ensuring that they are adequately funded and appropriately staffed. Administrative data, while not necessarily tailored to drug information needs, are utilized as indicators to inform and monitor drug policy. One example is the use of death certificate data; while such information is collected for many health-related purposes, death certificate data can provide insight into substance use related overdose deaths and patterns among such deaths. Such administrative data are repurposed to extract information on patterns and consequences pertaining to drugs. Many of our existing data sources originate from administrative records (see box below). Administrative data are rich but tend to be narrowly focused on the collecting agency’s mission and drug information is limited to their existing coding structures. For example, death certificate data use the International Classification of Diseases (ICD) standard developed by the World Health Organization to code all causes of death and includes a narrow range of codes specific to causes of death involving drugs. The need for more detail on specific drugs in death data will need to rely on additional information obtained from medical examiner or coroner (ME/C) reports. ME/C reports are a separate process Major Drug Data Sets Originating from Administrative Sources: Treatment Episode Data Set (TEDS) from reporting by treatment facilities receiving public funds to their single State agency (SSA), compiled by SAMHSA into a national dataset to provide characteristics of admissions to and discharges from substance abuse treatment. Death certificate data from States compiled and coded by the National Center for Health Statistics (NCHS), made available from Centers for Disease Control and Prevention’s (CDC) Wide-ranging Online Data for Epidemiologic Research (WONDER) database for data on drug deaths and the involvement of specific drugs. Healthcare Cost and Utilization Project (HCUP) from records of emergency department admissions and inpatient hospital stays from participating States compiled by the Agency for Healthcare Quality and Research (AHRQ) to provide data on drug overdoses and neonatal abstinence syndrome. National Forensic Laboratory 

1   2   3   4   5   6   7   8   9   10   11   12   13   14  15   16   17   18   19   20   21    22   23   24   25  26   27   28   29   30   31   32   33   34   35   36   37   38   39   40   41   42   43   44   45   46   47   48   49  50   51   52   53   54   55   56   57   58   59   60   61   62   63   64   65