Evidano home
Commentary on News

AI Qualitative Analysis of Hepatitis C Barriers

Evidano9 min read

Public health teams need to understand not only how many people face barriers to hepatitis C treatment, but also why those barriers delay care. The November 20, 2026 PLOS One study of Los Angeles County shows how mixed-methods research combines program data with patient interviews to reveal those reasons. This article refracts the study through the lens of AI-enabled qualitative research, focusing on how researchers can organize transcripts, compare treated and untreated groups, and connect themes to quantitative findings. The study itself used iterative human coding in Dedoose, so AI should be treated as an accelerant for synthesis rather than a substitute for methodological judgment, participant protection, or clinical expertise. Evidano is an AI-powered qualitative data analysis platform that helps public health and healthcare researchers analyze interviews, open-ended survey responses, and research documents. The result is a practical workflow for moving from raw narratives to defensible findings about language access, insurance, care navigation, stigma, and treatment motivation.

Key Takeaways

According to the PLOS One study on hepatitis C treatment access, 54% of 344 Los Angeles County case-residents contacted between October 2024 and January 2025 reported barriers to hepatitis C treatment. The study shows that qualitative analysis adds the patient context needed to explain how language barriers, cost, transportation, unstable housing, incarceration, and healthcare-system delays affect treatment access.

  • The November 20, 2026 PLOS One study found that personal barriers represented 43% of reported barriers among the 91 case-residents who specified a barrier, while provider-related barriers represented 31%.
  • The study analyzed interviews with 10 participants between October 2024 and January 2025, including 7 untreated participants and 3 treated participants.
  • The PLOS One researchers identified language support, patient navigation, care coordination, mobile clinics, treatment continuity after incarceration, and fewer insurance restrictions as potential responses.
  • AI-enabled qualitative analysis can help researchers compare themes across treatment status, language, insurance, and other segments, but human researchers must validate codes and protect participant confidentiality.
  • The study reported that only 30% of people diagnosed with hepatitis C in Los Angeles County between 2019 and 2023 knew about hepatitis C treatment, while 32% completed treatment, according to the background evidence summarized by the authors.

What the Hepatitis C Study Found

The study found that hepatitis C treatment access in Los Angeles County is shaped by interacting personal, provider, insurance, and healthcare-access barriers. The researchers used de-identified program data from 344 case-residents and thematic analysis of semi-structured interviews to answer both how often barriers occurred and how patients experienced them.

According to the November 20, 2026 PLOS One publication, 168 of the 344 case-residents had already received treatment and 176 had not received treatment when case managers contacted them between October 2024 and January 2025. Among participants who reported a specific barrier, personal barriers were most common at 43%, followed by provider-related barriers at 31%, insurance barriers at 13%, and inadequate access to medical care at 13%.

The qualitative sample added explanations that frequency data alone could not provide. Untreated participants described high treatment costs, language barriers, transportation problems, limited hepatitis C information, and difficulty navigating the healthcare system. Treated participants described homelessness, incarceration, drug addiction, and long waits for specialist care.

The researchers used an inductive coding approach, independently coded transcripts, and held team meetings to compare codes and resolve discrepancies. The study therefore demonstrates that AI can support qualitative synthesis, but an AI-generated theme should remain an auditable research aid rather than an unreviewed conclusion.

One untreated Spanish-speaking participant described the impact of communication barriers: “Since I don’t speak 100% English... I didn’t understand very well what they were telling me.” The participant also said that a new primary care doctor explained hepatitis C well in Spanish, according to the November 20, 2026 PLOS One study.

One treated participant described the effect of provider delays: “To start my medication, I took 2 months to meet the doctor.” The quotation shows why qualitative analysis is useful for distinguishing motivation to seek treatment from the system’s ability to deliver timely care.

Hepatitis C Treatment Barriers Snapshot

Date or periodMetricValueImplication for qualitative researchers
October 2024 to January 2025Case-residents contacted344A program dataset can establish the scale and distribution of reported barriers.
October 2024 to January 2025Case-residents already treated168, or 49%Treatment status supports cross-segment comparison between treated and untreated experiences.
October 2024 to January 2025Case-residents not treated176Untreated narratives can reveal why diagnosis does not become treatment initiation.
October 2024 to January 2025Case-residents reporting barriers54%More than half of the program population reported at least one access barrier.
October 2024 to January 2025Participants reporting specific barriers91A smaller analytic denominator requires careful interpretation of barrier percentages.
October 2024 to January 2025Interviews completed10, including 7 untreated and 3 treated participantsInterviews provide explanatory depth but are exploratory and not statistically generalizable.
2019 to 2023Los Angeles County hepatitis C diagnoses27, 678The background evidence indicates a substantial local population requiring linkage-to-care strategies.

Implications for AI-Enabled Public Health Research

AI-enabled qualitative research can help public health teams turn patient narratives into traceable intervention priorities, provided that researchers preserve context and review machine-generated outputs. The Los Angeles County study shows why a single overall barrier count is insufficient for designing equitable hepatitis C programs.

  • Segment by language: The PLOS One interviews suggest that Spanish-speaking untreated participants particularly valued provider guidance, so researchers should compare language-specific themes rather than treating language as a demographic footnote.
  • Separate individual from structural barriers: The study distinguishes motivation, knowledge, and stigma from provider delays, transportation, insurance, and access to care, which helps teams assign interventions to the correct level.
  • Compare treatment journeys: The study’s 7 untreated and 3 treated interview participants show how the same healthcare environment can produce both delay and successful engagement, making treated versus untreated comparison analytically valuable.
  • Connect frequency to meaning: The 43% personal-barrier figure identifies prevalence, while interview quotations explain how homelessness, addiction, cost, or unclear information can make treatment difficult.
  • Keep ethics visible: The study received USC Institutional Review Board approval and used verbal informed consent, while AI workflows should additionally use de-identification, access controls, and human review for sensitive health narratives.

A responsible AI workflow would begin with secure transcription and redaction, continue with researcher-defined codes and AI-assisted thematic grouping, and end with human review of quotations, contradictions, negative cases, and subgroup differences. Public health teams should not treat an automated frequency as proof of causation, especially when the PLOS One authors described the qualitative sample as exploratory and noted that small subgroup sizes prevented inferential analysis.

How Evidano Helps With Qualitative Analysis of Hepatitis C Treatment Barriers

Problem: Transcripts and program documents are difficult to synthesize consistently

Evidano helps researchers organize interview transcripts, research papers, reports, and other documents in one AI-assisted qualitative analysis workspace. Researchers can use thematic, content, frequency, and cross-segment analyses to examine barriers across treated and untreated participants or across English- and Spanish-language interviews.

Evidano’s qualitative research features support a workflow in which researchers inspect themes, review supporting passages, and retain responsibility for interpretation.

Problem: Multilingual interviews can hide important differences

Evidano provides transcription and translation workflows that can help teams process multilingual interview data while preserving research terminology through custom dictionaries. Researchers can compare translated themes with source-language excerpts before using quotations in a report.

Because the PLOS One study identified language barriers as a treatment-access theme, multilingual review is especially relevant for hepatitis C patient-experience research.

Problem: Researchers need to connect themes with participant segments

Evidano supports cross-segment analysis so researchers can examine how themes vary by treatment status, language, insurance category, or other study variables. Visualizations such as hierarchical codes and subcodes, word clouds, and co-occurrence networks can help teams inspect relationships among barriers and facilitators.

Researchers should use these outputs as analytical aids, then verify the underlying passages and sample sizes before making public health recommendations.

Problem: Sensitive health narratives require strong data governance

Evidano encrypts data and states that customer data is never used to train third-party models. Researchers working with hepatitis C interviews should also remove direct identifiers, limit access to approved team members, and follow institutional review board and informed-consent requirements.

Evidano’s data security information can help research teams evaluate platform controls before uploading sensitive qualitative data.

FAQ: Qualitative Analysis of Hepatitis C Treatment Barriers

What did the Los Angeles County hepatitis C study find?

The November 20, 2026 PLOS One study found that 54% of 344 contacted case-residents reported barriers to hepatitis C treatment. Personal barriers accounted for 43% of specified barriers, provider-related barriers for 31%, insurance barriers for 13%, and inadequate access to medical care for 13%.

Why were qualitative interviews important in this hepatitis C research?

Qualitative interviews explained why barriers occurred and how patients experienced them. The 10 interviews revealed context about language, cost, transportation, healthcare navigation, homelessness, incarceration, addiction, provider trust, and family support that program frequencies could not show by themselves.

How can AI support qualitative analysis in public health research?

AI can support public health qualitative analysis by accelerating transcription, translation, code organization, thematic synthesis, document comparison, and cross-segment analysis. Human researchers must still validate themes, inspect quotations, account for sample limitations, and protect participant privacy.

Did the hepatitis C study use artificial intelligence to code interviews?

The study reported using inductive coding and Dedoose Version 9.2.22 rather than describing an AI coding workflow. The researchers independently coded transcripts, compared codes, and resolved discrepancies through team discussion, which provides a useful standard for human oversight of AI-assisted analysis.

What interventions did the researchers recommend?

The researchers recommended language support, patient navigation, care coordination, mobile clinics for people experiencing homelessness, continuity of treatment after incarceration, and removal of insurance restrictions such as prior authorization. The recommendations were based on the study’s combined quantitative and qualitative findings.

Can these findings be generalized to every hepatitis C population?

The findings should be treated as exploratory rather than universally generalizable. The PLOS One authors noted that the qualitative sample included only 10 participants, was unevenly distributed between treated and untreated groups, and did not support inferential analysis for small barrier subcategories.

Conclusion and Next Steps

The Los Angeles County study shows that qualitative analysis is essential for understanding why hepatitis C treatment does not consistently follow diagnosis. AI-enabled tools can help public health researchers process transcripts, compare segments, and connect patient narratives with program statistics, but responsible research still depends on human coding review, transparent limitations, and strong privacy controls. Evidano brings these analysis tasks into one secure workflow for interviews, documents, and open-ended responses. Try Evidano for free to explore AI-assisted qualitative analysis for your next healthcare or public health research project.

Topics

  • qualitative analysis of hepatitis C treatment barriers
  • AI qualitative analysis
  • hepatitis C patient experience research
  • public health thematic analysis
  • mixed-methods healthcare research

Keep reading

Browse all articles