# Using data science to measure the impact of opioid agonist therapy in patients admitted with Staphylococcus aureus bloodstream infections

> **NIH NIH K08** · UNIVERSITY OF CALIFORNIA LOS ANGELES · 2021 · $203,040

## Abstract

PROJECT SUMMARY/ABSTRACT
This career development award will provide early career support for investigation of the management of infec-
tious diseases in the setting of addiction in hospitals. The award will provide support for the candidate to develop
expertise in the following areas: 1) addiction science research; 2) natural language processing; 3) machine learn-
ing; 4) professional development; and 5) responsible conduct of research. For this, Dr. Goodman-Meza will be
mentored by a multidisciplinary, cross-institutional team with expertise in addiction, infectious diseases, and data
science. His primary mentor, Dr. Steve Shoptaw, has an extensive track record in addiction-related research and
training of future independent investigators. His co-mentors include Dr. Alex Bui and Dr. Matthew B. Goetz. Dr.
Bui is an expert in biomedical data science and heads NIH training programs in this field. Dr. Goetz has broad
experience of productive infectious diseases clinical research within the Veterans Health Administration (VHA).
The current opioid epidemic in the United States has been associated with an increase in infections, in particular
hepatitis C and bacterial infections. Bacterial infections are the leading infectious diagnosis leading to hospitali-
zation in individuals with an opioid use disorder (OUD), and incur significant healthcare expenditures. Despite
the availability of opioid agonist therapy (OAT) in the form of methadone or buprenorphine, less than 20% of
people with OUD actually receive OAT. Hospitalization for a bacterial infection may be an ideal time to initiate
OAT, but the benefits of this practice are unknown. In this proposal, the candidate will assess the impact of
initiating OAT in people who inject opioids admitted to the VHA due to a Staphylococcus aureus blood stream
infection (bacteremia) – the most common bacterial pathogen among people who inject opioids. Using data
already collected for 36,868 cases of S. aureus bacteremia (SAB) from the VHA electronic data repository, the
candidate will address three research questions: 1) is a natural language processing algorithm (NLP) more ac-
curate than a standard International Classification of Diseases (ICD) code-based approach at screening records
to correctly identify individuals who inject opioids in a cohort of patients admitted with SAB; 2) what are the
temporal and geographic trends of SAB in people who inject opioids and those who receive OAT at the facility-
level; and 3) using a machine learning framework, what are the estimated impacts of OAT on patient centered
outcomes – death, readmissions, leaving against medical advice, and subsequent outpatient engagement in
OAT. These formative data will help the candidate to establish a productive early career as a physician-scientist
and advise development of an OAT-delivery strategy to mitigate infectious complications of injection opioid use.
Through this award, Dr. Goodman-Meza will establish himself as an expert physic...

## Key facts

- **NIH application ID:** 10164748
- **Project number:** 5K08DA048163-03
- **Recipient organization:** UNIVERSITY OF CALIFORNIA LOS ANGELES
- **Principal Investigator:** David Goodman
- **Activity code:** K08 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2021
- **Award amount:** $203,040
- **Award type:** 5
- **Project period:** 2019-06-15 → 2024-05-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10164748

## Citation

> US National Institutes of Health, RePORTER application 10164748, Using data science to measure the impact of opioid agonist therapy in patients admitted with Staphylococcus aureus bloodstream infections (5K08DA048163-03). Retrieved via AI Analytics 2026-05-23 from https://api.ai-analytics.org/grant/nih/10164748. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*