Dr. Travis Porco's research applies mathematical, statistical, and computational methods to problems in epidemiology, public health, clinical research, and evidence-based medicine. His work includes infectious-disease transmission and control, clinical trials, digital epidemiology, and the development and validation of new methods for extracting reliable scientific information from large and complex data sources.
UCSF Profiles ↗ Full publication list ↗
Research
A major current focus is the rigorous evaluation of large language models and related computational methods in biomedical research. Working with Thomas M. Lietman, Michael Deiner, Seth Blumberg and other Proctor collaborators, the group has developed methods for analyzing large collections of patient-generated health information, detecting epidemiologic signals in online data, and converting unstructured health text into structured measurements that can be evaluated against human experts and other reference standards.
The group is also developing methods for computational systematic evidence synthesis. Current work uses registered Cochrane review protocols to reconstruct systematic reviews through literature retrieval, screening, structured data extraction, statistical pooling, and evidence synthesis. A central goal is prospective validation: machine-generated reviews can be completed and recorded before the corresponding human reviews are published, providing a rigorous test of accuracy while reducing concerns about prior exposure to published answers. Controlled studies using expert-curated and deliberately modified source articles are used to determine where automated methods succeed, where they fail, and when human review remains necessary. This work builds on an existing protocol-driven pipeline developed by Porco and Deiner for prospective and retrospective validation of automated evidence synthesis.
These newer projects build on decades of work in mathematical epidemiology and clinical research, including studies of trachoma elimination, tuberculosis, measles, antimicrobial resistance, vaccine trials, and infectious and inflammatory eye disease.
That longer-running work has been carried out with the Lietman group at Proctor and international partners such as the Partnership for the Rapid Elimination of Trachoma (PI: Sheila West, Johns Hopkins), estimating the efficacy of mass azithromycin distribution for eliminating trachoma infection. Multi-year analyses of transmission dynamics in Tanzania found that transmission did not appear to intensify over time — helping to ease concerns that a loss of immunity following successful control could undermine future elimination efforts.
Research Themes
- Systematic evidence synthesis: development and prospective validation of computational methods for literature retrieval, screening, structured data extraction, meta-analysis, evidence appraisal, and reproducible systematic reviews
- Large language models in biomedical research: evaluation of LLMs against human experts, controlled ground truth, independent models, and prospective reference standards, with particular attention to reproducibility and model error
- Digital epidemiology and infodemiology: use of social media, internet search behavior, and other large-scale online data to study disease activity, symptoms, patient experiences, treatment concerns, and emerging public-health signals
- Infectious-disease modeling and surveillance: mathematical and statistical modeling of disease transmission, outbreak dynamics, intervention effectiveness, and population-level surveillance — including tuberculosis and measles
- Trachoma and global health: modeling elimination strategies, mass antibiotic distribution, transmission dynamics, and outcomes from large international clinical trials
- Antimicrobial resistance: quantitative studies of the emergence and population consequences of antimicrobial resistance, including resistance associated with population-level antibiotic interventions
- Clinical trials and biostatistical methods: design, analysis, simulation, and interpretation of randomized trials and other clinical studies, particularly in ophthalmology and infectious disease
Current and Past Lab Members
Selected Publications
- Deiner MS, Deiner RY, Fathy C, Deiner NA, Hristidis V, McLeod SD, Bukowski TJ, Doan T, Seitzman GD, Lietman TM, Porco TC. Use of Large Language Models to Classify Epidemiological Characteristics in Synthetic and Real-World Social Media Posts About Conjunctivitis Outbreaks: Infodemiology Study. Journal of Medical Internet Research. 2025. Evaluated LLM classification of epidemiologic characteristics using synthetic and real-world social-media reports with human-expert validation.
- Deiner MS, Honcharov V, Li J, Mackey TK, Porco TC, Sarkar U. Large Language Models Can Enable Inductive Thematic Analysis of a Social Media Corpus in a Single Prompt: Human Validation Study. JMIR Infodemiology. 2024;4:e59641.
- Deiner MS, Deiner NA, Hristidis V, McLeod SD, Doan T, Lietman TM, Porco TC. Use of Large Language Models to Assess the Likelihood of Epidemics From the Content of Tweets: Infodemiology Study. Journal of Medical Internet Research. 2024;26:e49139.
- Deiner MS, McLeod SD, Chodosh J, Oldenburg CE, Fathy CA, Lietman TM, Porco TC. Clinical Age-Specific Seasonal Conjunctivitis Patterns and Their Online Detection in Twitter, Blog, Forum, and Comment Social Media Posts. Investigative Ophthalmology & Visual Science. 2018;59(2):910–920.
- Aragie S, et al. Water, sanitation, and hygiene for control of trachoma in Ethiopia (WUHA): a cluster-randomised trial. Lancet Global Health. 2022;10(1):e87–e95.
- Arzika AM, et al. Biannual mass azithromycin distributions and malaria parasitemia in pre-school children in Niger (MORDOR). PLoS Medicine. 2019;16(6):e1002835.
- Blower SM, McLean AR, Porco TC, et al. The intrinsic transmission dynamics of tuberculosis epidemics. Nature Medicine. 1995;1(8):815–821.
- Blower SM, Porco TC, Darby G. Predicting and preventing the emergence of antiviral drug resistance in HSV-2. Nature Medicine. 1998;4(6):673–678.
- Banerjee R, Schecter GF, Flood J, Porco TC. Extensively drug-resistant tuberculosis: new strains, new challenges. Expert Review of Anti-infective Therapy. 2008;6(5):713–724.
- Blumberg S, Enanoria WT, Lloyd-Smith JO, Lietman TM, Porco TC. Identifying postelimination trends for the introduction and transmissibility of measles in the United States. American Journal of Epidemiology. 2014;179(11):1375–1382.
- Bellan SE, et al. Statistical power and validity of Ebola vaccine trials in Sierra Leone: a simulation study. Lancet Infectious Diseases. 2015;15(6):703–710.
- Acharya NR, et al. Stopping of adalimumab in juvenile idiopathic arthritis-associated uveitis (ADJUST): a randomised controlled trial. Lancet. 2025;405(10475):303–313.
- Acharya NR, et al. Outcomes in Patients With Vogt-Koyanagi-Harada Disease From the FAST Uveitis Trial. American Journal of Ophthalmology. 2024;267:100–111.
Education
- University of California, Berkeley, PhD, Biophysics (Advisor: Wayne Getz)
- University of California, Berkeley, MPH, Biostatistics
- University of California, San Francisco, Postdoctoral Scholar
- University of California, San Francisco, Fellow, Traineeship in AIDS Prevention Studies
Full publication list on PubMed ↗ Porco Phd All publication ↗ UCSF Profiles ↗