Skip to main content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.


The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Identifying Co-Occurring Disorders among Opioid Users Using Linked Hospital Care and Mortality Data: Capstone to an Existing FY18 OS-PCORTF Project

Improve Public Health Surveillance and Expand Researchers’ Access to Data on Health Outcomes of Opioid Users with Co-occurring Substance and Mental Health Issues
  • Centers for Disease Control and Prevention (CDC)
Start Date
  • 5/1/2019
  • Use of Enhanced Publically-Funded Data Systems for Research
  • Use of Clinical Data for Research
  • Linking of Clinical and Other Data for Research


STATUS: Completed Project


According to the 2017 National Survey on Drug Use and Health (NSDUH), the number of adults with substance use disorders (SUD) who had any mental illness was about 8.5 million and the number with severe mental illness was about 3.1 million people. Based on 2015 NSDUH data, approximately 1.5 million adults with severe mental illness had misused opioids in the past year, which is equivalent to a co-occurrence of opioid misuse and severe mental illness in an estimated 1 in 8 adults (13%). It is important for the National Hospital Care Survey (NCHS) to monitor the role that co-occurring disorders plays in opioid-related morbidity and mortality outcomes.

A previously funded OS-PCORTF FY18 project at NCHS titled, Enhancing Identification of Opioid-Involved Health Outcomes Using Linked Hospital Care and Mortality Data, has provided enhanced methodology to accurately identify a hospital patient’s use of opioids in any form (i.e., used as directed, misused in a manner contrary to provider instructions, used intentionally to become intoxicated or for the purpose of self-harm, taken accidentally, etc.). Additionally, the FY18 OS-PCORTF project identified the specific legal or illicit opioid agent taken. This OS-PCORTF FY19 Capstone project built upon the FY18 project methodology to flag evidence of co-occurring mental health disorders. Both projects used algorithms that determine the occurrence of an event (the use of opioids, type of opioid agent taken, and presence of a substance use or mental health issue) by selecting combinations of coded items (diagnoses, procedures, lab results, etc.) and terms contained in free-text (clinical notes, cause of death literal text). Both projects resulted in the creation of linked files that combine three data sources to enable access to data that follows patients with an opioid event for one year following hospital discharge. This allows for retrospective analysis of the extent to which specific opioid agents and the co-occurrence of mental health disorders were involved in hospital encounters preceding post-discharge deaths.


The goal of this project was to improve public health surveillance and expand researchers’ access to data on health outcomes of opioid users with co-occurring substance use and mental health issues.

Project Objectives:

  • Develop a new set of algorithms that uses the linked NHCS/ National Death Index (NDI)/ National Vital Statistics System restricted mortality data, drug specific information (NVSS-M-DO) files to identify hospital encounters and death records involving patients with co-occurring disorders using medical code-based algorithm and natural language processing.

  • Conduct a study to validate algorithms from this project and the FY18 Enhancing Identification project to identify the use of opioids and the existence of co-occurring disorders.

  • Apply the validated algorithm to identify prevalence of opioid-involved emergency department visits and co-occurring disorders among opioid users in the 2016 linked NHCS, NDI and NVSS-M-DO files.

  • Provide data on opioid use and co-occurring disorders and make that data available through: 1) the NCHS Research Data Center, and 2) a previously developed interactive web portal for NHCS participating hospitals.

  • Disseminate research findings from the validation study and the application of the validated algorithm to calculate prevalence of co-occurring disorders among opioid users in the linked data between NHCS, NDI, and the NVSS-M-DO files.


  • The project team developed and applied the co-occurring disorders algorithm, which utilizes coded medical data and natural language processing (NLP) methods to identify mentions of substance use disorders or mental health issues in the unstructured clinical data.

  • The team applied the co-occurring disorders algorithm to 2016 NHCS data to identify patients with opioid-involved hospitalizations and emergency department visits with co-occurring substance use disorders or mental health issues.

  • The project team completed a validation study to assess the performance of opioid use algorithms and co-occurring disorder algorithms.

  • The team linked the 2016 NHCS data to the 2016-2017 NDI and DIM data with enhanced information on opioid-involved hospital visits with co-occurring substance use disorders and mental health issues.



  • The team published the Identifying Co-Occurring Disorders among Opioid Users Final Report here:

  • The 2016 NHCS/NDI/DIM enhanced dataset is available through the Federal and NCHS Research Data Center. Information about how to request these data is available here:

  • The team published the “Identifying Co-Occurring Disorders among Opioid Users Using Linked Hospital Care and Mortality Data” report, which describes the completed 2016 NHCS/NDI/DIM enhanced dataset. The report is available here:

  • The Python code for the co-occurring disorders algorithm is available at CDC’s GitHub repository. The code can be found at the three links below:


    Below is a list of ASPE-funded PCORTF projects that are related to this project

    Enhancing Identification of Opioid-Involved Health Outcomes Using Linked Hospital Care and Mortality Data – National-level statistics on opioid-related hospitalizations are often incomplete. Electronic health record (EHR) data contain clinical notes and laboratory results, which allow a wider perspective on the hospitalization. This project aims to improve surveillance and expand researchers’ access to data on hospital care patterns and risk factors associated with opioid overdose deaths. To accomplish this, the project will merge the NHCS, NDI, and DIM. The linked data will support research examining characteristics of individuals who have opioid-related events, patterns of hospital use in months prior to death, and comparison of patients and services.

    Enhancing Data Resources for Researching Patterns of Mortality in Patient Centered Outcomes Research – Through collaboration between the CDC, Centers for Medicare and Medicaid (CMS), and Food and Drug Administration (FDA), the overall goal of this project is to increase the availability of information on the cause of death by linking NDI data to other sources. Enabling linkages will allow researchers to develop national estimates of cause-specific death rates following emergency department visits and/or hospital stays for specific conditions.

    Improving the Mortality Data Infrastructure for Patient-Centered Outcomes – Comprised of all U.S. mortality events since 1979, the NDI database allows researchers to match entries in the NDI to those participating in longitudinal clinical and epidemiologic studies to determine both fact and cause of death. A significant challenge with the NDI has been the lag between the date of death and the availability of the record for matching purposes. The CDC’s NCHS worked to improve the infrastructure to support more timely and complete mortality data collection through more timely delivery of state death records (e.g., cause of death) to the NDI database and through linking NDI records with nationally collected hospital datasets to obtain a more complete picture of patient care.