Hospital Discharge Data 

Last Updated: March 2022

Hospital discharge data are another source of information that can be used to understand drug overdose trends in the US. Traditionally, healthcare discharge data are collected for primarily administrative purposes, such as obtaining reimbursement for hospital care hospitals, but are also used for tracking utilization, costs, and quality of healthcare by jurisdiction health departments, hospitals, and hospital associations. These data provide details on the quality of healthcare services, types of services delivered, and billing charges on hospitalizations and ED visits. Moreover, the data present opportunities to survey injuries, including drug overdoses. Overdose-related ED visits or inpatient hospitalizations can be identified by querying any diagnosis field for International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) discharge diagnosis codes indicating a drug poisoning (i.e., T36-T50 codes). Although these data are not available as rapidly as other data sources (i.e., they are usually delayed several months to a year), they include standard data elements in the UB-04 form that may not be included in all electronic health record data. Therefore, they can be useful for understanding the burden of overdoses across states and localities over time.  See the ICD-10-CM Injury Surveillance Toolkit as a resource. 

The Agency for Healthcare Research and Quality (AHRQ) Healthcare Cost and Utilization Project (HCUP)¹ databases are derived from administrative data and contain encounter-level, clinical and nonclinical information including all-listed diagnoses and procedures, discharge status, patient demographics, and charges for all patients, regardless of payer (e.g., Medicare, Medicaid, private insurance, uninsured). Several relevant databases for monitoring drug overdose include both state and national inpatient data (i.e., State Inpatient Database [SID] and Nationwide Inpatient Sample [NIS]) and state and national emergency department data (i.e., State Emergency Department Database [SEDD] and Nationwide Emergency Department Sample [NEDS]). 

All but two states (Alabama and Idaho) and the District of Columbia maintain statewide hospital discharge data systems and report that data to HCUP for inclusion in the SID and NIS and 42 states share with the SEDD and NEDS. Most HCUP data is available approximately two years after receipt and could be useful for analysis of overdose trends over time but does not provide current information about overdose events. 

Several commercial sources also provide de-identified claims data including IQVIA, Optum, and MarketScan.  IQVIA, for example, provides institutional claims data in their Hx product on a daily to weekly schedule and represents 60 million patients. Data elements in the Hx product include gender, age, three-digit zip code, patient’s diagnosis, procedures, length of stay, and drugs administered and are sourced from the UB-04 or EDI 837I healthcare claim transactions. 

The National Hospital Care Survey (NHCS) is a national probability sample survey that describes national patterns of health care delivery in hospital-based settings, including inpatient departments, emergency departments (EDs), and outpatient departments (OPD). NHCS integrates the data formerly collected by the National Hospital Discharge Survey (NHDS) with the ED, OPD, and ambulatory surgery center data collected by the National Hospital Ambulatory Medical Care Survey (NHAMCS). Data elements collected in NHCS include demographics, diagnoses, procedures, laboratory tests, and medications, along with patient identifiers that allow for linking of inpatient and outpatient records and outside databases, such as the National Death Index. NHCS uses electronic health records (EHRs) and the UB-04 standard billing file as data sources. 

¹Available at https://www.ahrq.gov/data/hcup/index.html