Opioid-Related Hospital Use

Inpatient stays and emergency department (ED) visits including opioid-related hospital use are identified by any diagnosis (all-listed) in the following ranges of International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes:

• 304.00-304.02: Opioid type dependence (unspecified; continuous; episodic).
• 304.70-304.72: Combinations of opioid type drug with any other drug dependence (unspecified; continuous; episodic).
• 305.50-305.52: Opioid abuse (unspecified; continuous; episodic).
• 965.00-965.02; 965.09: Poisoning by opium (alkaloids), unspecified; heroin; methadone; other opiates and related narcotics.
• 970.1: Poisoning by opiate antagonists.
• E850.0-E850.2: Accidental poisoning by heroin; methadone; other opiates and related narcotics.
• E935.0-E935.2: Heroin, methadone, other opiates and related narcotics causing adverse effects in therapeutic use.
• E940.1: Opiate antagonists causing adverse effects in therapeutic use.

It should be noted that ICD-9-CM diagnosis codes related to opioid dependence or abuse "in remission" are not used to identify opioid-related hospital use because remission does not indicate active use of opioids.

Unit of Analysis

The unit of analysis is the hospital discharge (i.e., the hospital stay) or an emergency department (ED) visit, not a person or patient. This means that a person who is admitted to the hospital or visits the ED multiple times in one year is counted each time as a separate "discharge" from the hospital or a separate "encounter" in the ED. Transfers to another acute care hospital are excluded.

Rate of Stays or Rate of ED Visits per 100,000 Population

The rate of inpatient stays or rate of ED visits includes the HCUP number of stays or ED visits in the numerator and the U.S. resident population in the denominator (with a multiplier of 100,000). The rate itself is provided as an unrounded value. The denominator is consistently defined with the numerator (i.e., rates for females use HCUP counts and population counts specific to females). Population data are obtained from the Nielsen Company, a vendor that compiles and adds value to data from the U.S. Bureau of Census. Nielsen uses intercensal methods to estimate annual household and demographic statistics for geographic areas.

The number of years of data reported for each individual State and the United States depends on the availability of the underlying HCUP database. For example, the HCUP nationwide databases for the most recent data year can only be created after all of the necessary State databases are available. State-level data are included in Fast Stats when they become available.

The discharge/visit counts are available in the exported data file, which can be downloaded by expanding "Show Data Export Options." Counts are rounded to the nearest 50 discharges or visits, with any counts less than 26 suppressed for confidentiality.

Inpatient Stays

State-level statistics on inpatient stays are from the HCUP State Inpatient Databases (SID). The SID are limited to patients treated in community hospitals in the State. Community hospitals are defined as short-term, non-Federal, general, and other hospitals, excluding hospital units of other institutions (e.g., prisons). Included among community hospitals are obstetrics and gynecology, otolaryngology, orthopedic, cancer, pediatric, public, and academic medical hospitals. Excluded are community hospitals that are also long-term care facilities such as rehabilitation, psychiatric, and alcoholism and chemical dependency hospitals.

We adjust the discharge counts for hospitals that were not included in the SID. Across all States, the SID are missing about 7 percent of community hospitals and about 1.5 percent of discharges. Weighting for missing hospitals uses the following information from the American Hospital Association (AHA) Annual Survey of Hospitals to define strata within the State:

• Ownership (government, private nonprofit, and private investor-owned)
• Size of the hospital based on the number of beds (small, medium, and large categories defined within region)
• Location combined with teaching status (rural, urban nonteaching, urban teaching).

If a stratum is missing one or more hospitals in the State data, then we set the discharge weight to the total number of discharges reported in the AHA divided by the total number of discharges in the State data. If all hospitals in a stratum are represented in the State data, then we set the discharge weight to 1. We also adjust the discharge weights for hospitals that have missing discharge quarters of data, provided there is no indication in the AHA Annual Survey that the facility had closed.

Discharge weights are specific to the data year for SID through 2013 (e.g., discharge weights for the 2013 SID use 2013 AHA data). Weighting for HCUP data starting in 2014 is based on AHA data from the prior year because current information is often unavailable (e.g., discharge weights for the 2014 SID use 2013 AHA data).

National statistics on inpatient stays are from the HCUP National (Nationwide) Inpatient Sample (NIS). The NIS is sampled from the HCUP State Inpatient Databases (SID). Beginning with the 2012 data year, the NIS is a 20 percent sample of discharges from community hospitals, excluding rehabilitation and long-term acute care (LTAC) hospitals, participating in HCUP in that data year. For data years 1988 through 2011, the NIS was a 20 percent sample of community, nonrehabilitation hospitals and included all discharges within sampled hospitals. The national estimates on inpatient stays presented in this section of Fast Stats were developed using the NIS Trend Weight Files for consistent estimates across all data years (e.g., LTACs were removed from earlier data years using trend weights).

Emergency Department Visits

Emergency department (ED) visits are defined as ED encounters that do not result in a hospital admission to the same hospital (i.e., treat-and-release ED visits).

State-level statistics on ED treat-and-release visits are from the HCUP State Emergency Department Databases (SEDD). The SEDD is limited to patients treated in community hospital-owned EDs in the State.

We adjust the ED visit counts for hospital-owned EDs that are missing from the SEDD. Across all States, the SEDD are missing about 5 percent of EDs and about 2.0 percent of ED visits. Data from the following data sources are used to weight for missing information: the American Hospital Association (AHA) Survey of Hospitals and the Trauma Information Exchange Program (TIEP) database, a national inventory of trauma centers in the United States collected by the American Trauma Society. Weighting for missing EDs uses the following information to define strata within the State:

• Ownership: government, private nonprofit, and private investor-owned (AHA)
• Location: large metropolitan, small metropolitan, micropolitan, and rural (AHA)
• Teaching status: nonteaching and teaching (AHA)
• Trauma center designation: levels I, II, and III (TIEP).

If a stratum is missing one or more EDs in the State data, then we set the weight to the total number of ED visits reported in the AHA divided by the total number of ED visits in the State data. If all EDs in a stratum are represented in the State data, then we set the discharge weight to 1. We also adjust the discharge weights for EDs that have missing quarters of data, provided there is no indication in the AHA Annual Survey that the facility had closed.

Discharge weights are specific to the data year for ED visits through 2013 (e.g., discharge weights for the 2013 ED visits use 2013 AHA data). Weighting of HCUP data for ED visits starting in 2014 is based on AHA data from the prior year because current information is often unavailable (e.g., discharge weights for the 2014 ED visits use 2013 AHA data).

National statistics on ED treat-and-release visits are from the HCUP Nationwide Emergency Department Sample (NEDS). Treat-and-release records were selected from the NEDS using the HCUP data element HCUPFILE, which identifies the source of the ED record: the HCUP State Emergency Department Databases (SEDD) or the HCUP State Inpatient Databases (SID). All records where HCUPFILE was equal to SEDD are included in this analysis, that is, inpatient admissions from the ED were excluded since these cases are represented in the NIS.

Age

Age refers to the age of the patient at admission. Discharges or visits missing age are excluded from results reported by age.

Sex

All nonmale, nonfemale responses are set to missing. Discharges or visits with missing values for sex are excluded from results reported by sex.

Community-Level Income

Community-level income is based on the median household income of the patient's ZIP Code of residence. Quartiles are defined so that the total U.S. population is evenly distributed. The cut-offs for the quartile designation are determined annually using ZIP Code demographic data obtained from the Nielsen Company, a vendor that compiles and adds value to data from the U.S. Bureau of Census. Nielsen uses intercensal methods to estimate annual household and demographic statistics for geographic areas. The value ranges for the national income quartiles vary by year. Income quartile is missing if the patient is homeless or foreign. Discharges missing the income quartile are excluded from results reported by community-level income.

Patient Location

Fast Stats includes five categories for patient location: Large central metropolitan, large fringe metropolitan (suburbs), medium metropolitan, small metropolitan, and rural. This is based on the six-category, county-level scheme developed by the National Center for Health Statistics (NCHS) to study the relationship between urbanization and health. In the NCHS scheme, the rural counties are divided into micropolitan and noncore categories, but in this section of Fast Stats, these are combined into a single category in order to preserve results when cell sizes are too small. For rates prior to 2014, the NCHS classification is based on population density from the 2000 Census. Starting in 2014, the NCHS classification is based on population density from the 2010 Census.

In the 2007 Rhode Island SID, the reporting of patients residing in counties designated as large central metropolitan was inconsistent with prior and subsequent years. Therefore, the fluctuation between 2006 and 2008 in the inpatient rates for opioid-related inpatient stays by urban-rural location should be considered an anomaly.